microfluidics technology for manipulation and analysis of biological

Analytica Chimica Acta 560 (2006) 1–23

Review

Microfluidics technology for manipulation and analysis of biological cells

Changqing Yi, Cheuk-Wing Li, Shenglin Ji, Mengsu Yang∗Department of Biology and Chemistry, City University of Hong Kong, 83 Tat Chee Avenue, Kowloon, Hong Kong, China

Received 4 October 2005; received in revised form 7 December 2005; accepted 12 December 2005Available online 25 January 2006

Abstract

Analysis of the profiles and dynamics of molecular components and sub-cellular structures in living cells using microfluidic devices has becomea major branch of bioanalytical chemistry during the past decades. Microfluidic systems have shown unique advantages in performing analyticalfunctions such as controlled transportation, immobilization, and manipulation of biological molecules and cells, as well as separation, mixing, anddilution of chemical reagents, which enables the analysis of intracellular parameters and detection of cell metabolites, even on a single-cell level.This article provides an in-depth review on the applications of microfluidic devices for cell-based assays in recent years (2002–2005). Variouscell manipulation methods for microfluidic applications, based on magnetic, optical, mechanical, and electrical principles, are described withselected examples of microfluidic devices for cell-based analysis. Microfluidic devices for cell treatment, including cell lysis, cell culture, and

devices for

. . . 2. . 2. 2. 3. 4. 6. . 7. . 7. . 7. . 810

. 11. 11. 11

1417

. 18

. 19

. 20. 20

cell electroporation, are surveyed and their unique features are introduced. Special attention is devoted to a number of microfluidiccell-based assays, including micro cytometer, microfluidic chemical cytometry, biochemical sensing chip, and whole cell sensing chip.© 2005 Elsevier B.V. All rights reserved.

Keywords: Microfluidic devices; Lab-on-a-chip; Miniaturized total analysis system; Biochip; Cell analysis; Cell manipulation

Contents

1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2. Cell manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

2.1. Magnetic manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.2. Optical manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.3. Mechanical manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.4. Electrical manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .2.5. Other manipulation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

3. Cell treatment. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.1. Cell lysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.2. Cell culture. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .3.3. Electroporation, electrofusion, and optoporation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

4. Cell analysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.1. Micro cytometer. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.2. Chemical cytometry. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.3. Biochemical sensing. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.4. Electrical characterization and ion channel studies. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .4.5. Whole cell assay. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

5. Concluding remarks. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Acknowledgement. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .Reference. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .

∗ Corresponding author. Tel.: +86 27 886945; fax: +86 27 886554.E-mail address: [email protected] (M. Yang).

0003-2670/$ – see front matter © 2005 Elsevier B.V. All rights reserved.doi:10.1016/j.aca.2005.12.037

2 C. Yi et al. / Analytica Chimica Acta 560 (2006) 1–23

1. Introduction

Micro total analysis systems (�TAS), also called “lab-on-a-chip”, integrate analytical processes for sequential opera-tions like sampling, sample pre-treatment, analytical separation,chemical reaction, analyte detection, and data analysis in a sin-gle microfluidic device. Microfluidic-based research has madesignificant advances over the last few years, and become verymuch a ‘hot topic’ recently. Because of their advantages includ-ing low reagent and power consumption, short reaction time,portability for in situ use, low cost, versatility in design, andpotentials for parallel operation and for integration with otherminiaturized devices, microfluidic chip-based systems for bio-logical cell studies have attracted significant attention. Microflu-idic technique has started to play an increasingly important rolein discoveries in cell biology, neurobiology, pharmacology, andtissue engineering.

As cell-based assay is deemed to be essential for thefunctional characterization and detection of drugs, pathogens,toxicants, and odorants, several review articles concerningmicrofluidics for cell analysis have been published during thepast few years. Berg and Andersson[1] presented a comprehen-sive review of microfluidics for cellomics, which covered themicrofluidic devices for cell sampling, cell trapping and sorting,cell treatment, and cell analysis. Erickson and Li et al.[2]focused on integrated microfluidic devices for cell handling andc ting,ap ted om storas ings tionrb ep-a alsor icd ingo elsaW s ofu tingb sedm wc syst ions rtingsp gb es tb

idicd endst idicd rdingt isd , cell

treatment, and cell analysis. Based on different physical forcesemployed, microfluidic devices for cell manipulation can becategorized into magnetic manipulation, optical manipulation,mechanical manipulation, and electrical manipulation. Selectedexamples for each manipulation method are described in detailand compared to each other to reveal their advantages and disad-vantages. Microfluidic devices for cell treatment, which includescell lysis, cell culture and cell electroporation, electrofusion, andoptoporation, are also surveyed. Special attention is devoted to anumber of microfluidic devices for cell-based analysis which canbe further categorized into micro cytometer, chemical cytome-try, biochemical sensing, and whole cell sensing, according totheir different applications. Based on these reviews, an outlookof this promising and rapidly expanding area will be presentedin the concluding remarks.

2. Cell manipulation

Fabrication of miniaturized cell manipulator is the first steptowards microfluidics for cell-based assays. The commonly usedmethods for cell manipulation on chip can be categorized basedon the manipulating force employed. In the following sections,a few selected microfluidic devices employing different forcesfor cell manipulation are described.

2.1. Magnetic manipulation

ares d forc m-m diam-e le ont andc iousm uently,c tropyi d effi-c iningrt ncen-t am s andBt s arect nineo t. Int neticw ge ofe s at anaW ew odsf atedb canb ntaln en

ytometry, dielectrophoretic cellular manipulation and sornd general cellular analysis. Manz and co-workers[3–5]resented a series of comprehensive reviews concentraicro total analysis systems which covered development hind theory of miniaturization[3], fabrication of microfluidicystems[3,4], and microfluidic standard operations includample preparation, sample injection, sample manipulaeaction, separation, and detection[4,5]. Microfluidic-basediological applications such as cell culture, PCR, DNA sration, DNA sequencing, and clinical diagnostics wereeviewed[4,5]. Optical manipulation of cells in microfluidevices, including the parallel manipulation of cells usptical tweezers, optical switching of cells in fluidic channnd optical handle had been reviewed by Ozkan et al.[6].hitesides and co-workers[7] emphasized on the advantage

sing PDMS for miniaturized bioassays, including cell sory flow cytometry and magnetic sorting on PDMS-baicrofluidics. Recently, microfluidics for cell culture, flo

ytometers, and other microscale flow-based cell analysisems was reviewed[8,9], where cell detection and enumeratystems and microfluidic fluorescence-activated cell soystems were detailedly described[9]. In the review articleresented by Toner and Irimia[10], principles for manipulatinlood cells at microscale and high-throughput approachlood cell separation using microdevices was introduced.

There are a large number of newly reported microfluevices for cell research in last 3 years. The article int

o provide an in-depth look at the applications of microfluevices for cell analysis in recent years (2002–2005). Acco

o the different functions of microfluidic devices, this reviewivided into three main sections, covering cell manipulation

ny

,

-

o

Magnetic manipulation, in which magnetic particleselectively attached to cells, is a commonly used methoell separation or purification in microfluidic device. The coonly used super-paramagnetic beads are 10–100 nm inter, whose extremely small size makes them very gent

he cells with apparently no effect on the cellular functionell viability. The magnetic field gradients generated by varethods can be employed to capture the bead and, conseq

ell attached to the bead, because of the magnetic anisonherent in the paramagnetic beads. The high specificity aniency of magnetic methods make them quite useful in obtaare cell types, especially in handling red blood cells[10]. Duringhe last few years, there are several research reports corating on this method[11–16]. Berger et al.[11] developedagnetic sorting device which used asymmetry in obstaclerownian motion to affect separation of objects.Fig. 1A shows

he basic schematic of the microdevice: ferromagnetic wireountersunk into the bottom of the central chamber at a 45◦ angleo the flow of cells and more than 350 channels feed intoutlets where unlabeled cells will flow into the central outle

his microdevice, cells are forced to move along the magires because the high magnetic field gradient at the edach wire imposes a force on the superparmagnetic beadngle to the hydrodynamic force (illustrated inFig. 1B) [11].estervelt and co-workers[12] combined microfluidic devicith microelectromagnetic matrix to develop various meth

or manipulating biological cells. The magnetic field genery the matrix can stably control individual cells in fluid ande configured dynamically to meet a variety of experimeeeds. Furdui and Harrison[13] comparatively evaluated op

C. Yi et al. / Analytica Chimica Acta 560 (2006) 1–23 3

Fig. 1. (A) The basic schematic of the microfabricated cell fractionation chip.Inset: SEM image of magnetic wires countersunk into the floor of the central chamber.(B) The magnetic force separation idea. High magnetic field gradients provide forces at an angle to the flow of cells. Reproduced with the permission from [11], ©2001 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim.

tubular capture beds with magnetically trapped bead for the trap-ping and enrichment of relatively rare cells in blood samples. Themagnetic bead bed approach provides more capture surface areaand places cells and surfaces in closer proximity, and allows thecells to be readily released again after washing for further sam-ple processing. An integrated microfluidic device consisting of achaotic mixer, an incubation channel, and a capture channel wasdeveloped for magnetically separatingE. coli with the captureefficiency of 53 and 37% from PBS and whole blood, respec-tively [14]. Inglis et al.[15] incorporated an array of microfab-ricated magnetic stripes into a microfluidic device to achievecell by cell separation using super-paramagnetic nanoparticles.Frazier and co-workers[16] achieved magnetophoretic separa-tion of red and white blood cells from whole blood based ontheir native magnetic properties. No additives such as magnetictagging or inducing agents were needed for this separation. Mag-netic method is a clean, versatile, and non-invasive method forcell manipulation in microfluidics. With the generation of newmagnetic bead materials and modification methods for conju-gating various ligands to magnetic beads, it is believed that themagnetic manipulation method will become more efficient andeasily integrated with microfluidics for cell-based assay.

2.2. Optical manipulation

Recently, there is an increasing interest in optical manip-u toi cessI icalm res

the focal point of an electric field gradient. Due to its inher-ent advantages, optical tweezers can be easily combined withmicrofluidic systems and is the most commonly used opticalmanipulation method in microfluidic devices. Umehara et al.[17] employed optical tweezers to position bacteria in a single-cell microcultivation assay where single bacterial cells couldbe isolated and positioned in microchambers. In this way, thereproducibility of genetically identical bacteria can be studiedby flowing different nutrients through the microfluidic systems.Hanstorp and co-workers[18] developed a microfluidic systemwhich employed optical tweezers to manipulate cells withoutthe media being dragged along with them within seconds, andevaluated it by exposingE. coli to different fluorescent markers.Optical tweezers offer high resolution for single cells trapping,but have limited manipulation area owing to tight focusingrequirements[6]. In order to improve the throughput, Flynn etal. [19,20]demonstrated the use of individually addressable ver-tical cavity surface emitting lasers (VCSEL), where each laserin the array was focused and acted as an individual trappingand manipulation source, as illustrated inFig. 2, for opticaltrapping and active manipulation of living cells and micro-spheres. The high device density and compactness achievablewith VCSEL arrays make them suitable for optical manipulationapplications, where individual high-throughput manipulation ofcells is necessary[19,20]. However, VCSELs have a relativelylow output power, and hence a lower trapping strength[6].R pat-t rkers[ izedd resis

lation of biological species on microfluidic device duets non-contact and contamination-free manipulation prot is well known that optical tweezers could tether biologolecules to dielectric spheres and then capture the sphe

.

at

ecently, “lab on a microscope”, which uses light intensityern to manipulate cells, has been presented. Wu and co-wo21] developed optoelectronic tweezers (OET), which utilirect optical image to create high-resolution dielectropho


Fig. 2. VCSEL array optical tweezers for the parallel transport of samples ona chip. Reproduced with the permission from[6], © 2003 Kluwer AcademicPublishers.

(DEP) electrodes for parallel manipulation of single particles,to achieve high resolution and high throughput at the same time.As illustrated inFig. 3, an upper ITO-coated glass and a lowerphotoconductive surface are biased with an ac signal, and liq-uid containing cells of interest is sandwiched between them.When projected light illuminates the photoconductive layer, thevirtual electrodes are created, and consequently forming non-uniform electric fields which enable particle manipulation viaDEP forces. MacDonald et al.[22] exploited the optical polariz-

FlcalwDG

ability of particles with an extended, three-dimensional opticallattice for on-chip cell sorting and achieved the sorting efficiencyof ∼100%. Though optical manipulation can provide variousadvantages, its applications in microfluidics for cell-based assayare still limited due to the complicated optical setup, complexoperation, and expensive instrumentation.

2.3. Mechanical manipulation

Due to the complex physical properties of biological cells,manipulating cells mechanically on a microfluidic chip posescertain challenges. Separation of target cells on microfluidicstructures for culture and assay purposes is the main applica-tion of on-chip mechanical manipulation. Separation of cellswithin the microchannels can be achieved through fabricatingconstriction structures such as microfilters[23–26], microwells[27–29], microgripper[30], dam structure[31,32], and sandbagstructure[33], or modifying microchannel interior surface withreactive coatings[34–36], with antibodies[37,38], with selectin[39] and with enzymes[40].

Size-dependent filter-based microfluidic devices have beenfabricated to concentrate microbial cells for immunofluorescentassay. Target cells with a fluorescent signal-to-noise ratio of 12could be observed at single-cell level using a 10–100 time moredilute staining solution than conventional concentration within2–5 min[23]. Four successively narrower regions of channelsw a-r lood[ ee aratet f ane ationb ationb -i etricb ze isr frac-t ticlesa -s ells( ee andc -8-b es ith-o idicd ascT ratedb the

ig. 3. The scheme of optoelectronic tweezers. An upper ITO-coated glass andower photoconductive surface are biased with an ac signal, and liquid containinells of interest is sandwiched between them. A light-emitting diode operatingt a wavelength of 625 nm is employed as illumination source. When projected

ight illuminates the photoconductive layer, the virtual electrodes are turned onhich can create non-uniform electric fields and enable particle manipulation viaEP forces. Reprinted with the permission from[21], © 2005 Nature Publishingroup.

v -t k ind n thep ed byfl siso za-t sions

ag

,

ith microfilters (as shown inFig. 4A) were constructed to sepate cultured neuroblastoma cells when mixed with whole b24]. Moorthy and Beebe[25] developed a microfluidic devicquipped with porous filters inside a microchannel to sep

he blood cells from serum in real-time, without the need oxternal power source. The efficiency of cell/serum separy the porous filter was found to be comparable to separy centrifuge. Sturm and co-workers[26] reported a microflu

dic particle-separation device that made use of the asymmifurcation of laminar flow around obstacles. As the gap sieduced using nanofabrication, this device can be used toionate biological supramacromolecules, such as viral parnd protein complexes. Langer and co-workers[27] used a twotep process to fabricate poly(ethylene glycol) (PEG) microwas shown inFig. 4B) within microchannels which could bmployed to dock cells within pre-defined locations. Leeo-workers[30] developed a microfluidic device with a SUased microgripper (as shown inFig. 4C) that could manipulatingle cells in physiological ionic solutions and hold a cell wut exerting any forces in the closed position. A microfluevice with a dam structure in parallel to the fluid flow wonstructed for docking and alignment of biological cells[31].he structure is composed of two parallel channels sepay a dam, where the height of the dam is smaller thanertical height of the channels (as shown inFig. 4D). The strucure allows cells to move in microfluidic channels and docesired locations with controllable number. Cells docked oarallel dam structure are exposed to minimal stress causuidic pressure[31]. A comprehensive hydrodynamic analyf two different types of dam structures for cell immobili

ion were carried out to compare the effects of dam dimen


Fig. 4. (A) SEM image of microfilters showing the transition from the 15-�m wide channels to the 10-�m-wide channels. Reprinted with the permission from[24],© 2004 IEEE. (B) SEM image of microwell structure. Reprinted with the permission from[24], © 2004 Royal Society of Chemistry. (C) SEM and close-ups of theoverhanging fabricated SU-8 microgripper. Reprinted with the permission from[30], © 2005 IEEE. (D) Cross section of the microchip showing the dimensions ofthe channels and the dam structure. Reprinted with the permission from[31], © 2002 American Chemical Society. (E) Photographic image of an array of sandbagsproduced on the PDMS substrate. Reprinted with the permission from[33], © 2003 Royal Society of Chemistry.

and cell docking on the pressure and velocity of the flow inthe microchannels as well as the hydrodynamic force and shearstress on the docked cells. The results shows that the paralleldam provides a better structure for cell docking than the perpen-dicular dam and the hydrodynamic force and shear stress on thecells can be decreased by adjusting the pressure in the auxiliarychannel of the parallel dam without significantly affecting theflow velocity and transportation efficiency[32]. Sandbag struc-tures (as shown inFig. 4E) were also developed to immobilizecells using single step photolithography and it was demon-strated that multi-height sandbag structure could be appliedfor cellular analysis using immunocytochemical staining assays[33].

Langer and co-workers[34] developed polymer-basedmicrofluidic devices with submicrometer thin reactive coat-ings deposited on the interior surface to provide defined andchemically reactive interfaces for biomolecules immobilization.Modification of cell adhesion on microchannels can also beachieved by coating silica microchannel surface with polyacry-lamide films[35] and octadecyltrimethoxysilane (OTMS) orN-(trieth-oxysilylpropyl)-O-polyethylene oxide urethane (TESP)[36]. Microfluidic cell-affinity chromatography (CAC) systemuses surface-immobilized antibodies for cell separation. Tonerand co-workers[37] achieved the separation of highly purepopulations of T and B lymphocytes from mixtures without pre-processing incubation using microfluidic chambers coated witha ition

on the cell separation, Toner and co-workers[38] developed acytometry platform which could create high-density leukocytearrays for rapid optical characterization. Laser-mediated sortingof cells was also demonstrated by retrieving single selected T-lymphocytes from the cell array using laser capture microdissec-tion (LCM) technology. Chang et al.[39] utilized selectin-basedtransient cell adhesions to capture, enrich, and even fraction-ate different types of cells in a microstructured fluidic channel.Payne and co-workers[40] reported an enzyme-based methodto in situ entrap, grow, and release cells under mild conditionswithin a biopolymeric hydrogel matrix.

Size-dependent filter-based microfluidic devices exhibitnumerous advantages, such as high labeling efficiency, shortdetection time, high reproducibility based on simple and robustexperimental procedures, high detection sensitivity at a wholecell level. However, their applications for cell-based assay arelimited by their poor selectivity. Other constriction structuresalso face the same challenge. Microgripper can precisely manip-ulate single cell without any damage, but the complicated fabri-cation procedures make it difficult to be applicable in commonbiology and chemistry labs. Oppositely, dam and sandbag struc-tures can be easily fabricated using various methods and canalso efficiently immobilize cells with minimal stress on them.Dam and sandbag structures may have more and more applica-tions in future microfluidic cell analysis. Although the coatingmethod which use cell adhesion to capture cells suffers fromr nd
ntibodies. Based on the effects of flow and surface cond selatively low immobilization efficiency, its high selectivity a


easily modifying microchannel interior surface still make it apromising method for developing microfluidic devices for cell-based assay.

2.4. Electrical manipulation

Dielectrophoresis (DEP) is the electronic analog of opticaltweezers[17,18] and is defined as the lateral motion impartedon uncharged particles as a result of polarization induced bynon-uniform electric fields. Scheme of two types of microfluidicDEP trap is shown inFig. 5. The standard way to make a DEPtrap is to create an electric field gradient with an arrangementof planar metallic electrodes (Fig. 5A) either directly connectedto a voltage source or free-floating in presence of an ac field[41]. One advantage of DEP is that it can be scaled for parallelelectrical manipulation of micro- and even nano-sized objectssuch as cells, DNA, proteins, nanoparticles, and possibly sin-gle molecules in aqueous solutions. Alternatively, a constrictionor channel in an insulating material can also be employed tosqueeze the electric field in a conducting solution. This elec-trodeless DEP (EDEP) (as shown inFig. 5B and C) providesvarious advantages over DEP, such as a very high electric fieldmay be applied without gas evolution as well as a huge increasein the dielectric response at low frequencies (below 1 kHz)[41].DEP has been successfully applied on microfluidic devices tomanipulate a variety of biological cells, such as bacterial, yeast,a

yso y thei tion,s reticf flec-t ofD pula-ta on-fi Them sys-t thati sedo n inF dedq maticrpt fieldc con-fi eds d thet sym-m owerst tingtc elec-t om-pd nism

Fig. 5. Scheme of a microfluidic DEP trap. (A) A metallic DEP trap made ofmicrofabricated wires on a substrate. The wires may be either free-floating orconnected to a voltage source. (B) An electrodeless DEP trap made of dielectricconstrictions. The solid lines are electric field lines E. (C) A scanning electronmicrograph of an electrodeless DEP device consisted of a constriction arrayetched in quartz. The constrictions are 1�m wide and 1.25�m deep. The wholechip measures 1 cm× 1 cm. The applied electric field directionz is shown bythe double-headed arrow. Reprinted with the permission from[41], © 2002HighWire Press.

from a large number of microbes in microfluidic device by usingboth a laser-trapping force and a dielectrophoretic force. A lin-ear traveling wave dielectrophoretic (twDEP) microchip wasdeveloped by Cui et al.[47] for cell separation. Viable cells canbe separated from heat-treatedListeria innocua cells on micro-

nd mammalian cells[42–50].Durr et al. [42] used a microfluidic chip with 3D arra

f microelectrodes embedded in microchannels to studnfluence of channel height, particle size, buffer compositrength, and frequency of electric field on the dielectrophoorce, and also the effectiveness of dielectrophoretic deion structures. Huang et al.[43] demonstrated the potentialEP for accurate genetic analysis of specific cell subpo

ions in heterogeneous samples. Voldman et al.[44] developedmicrofluidic cytometer where DEP was employed to c

ne cells and hold them against disrupting fluid flows.icro cytometer consisted of a cell array chip, an optical

em luminescently interrogating cells, and a control systemmplemented sorting function. Cell array chip was compof a regular array of non-contact single-cell traps (as showig. 6B and C) which was designed with asymmetric extruuadrupole geometry and electrically addressable. Scheepresentation of operation is depicted inFig. 6A. Switching theotential on the controlling electrode from +V to−V disrupts

he quadrupolar potential energy by removing the electricage, thus ejecting the cell. Particles can be more stronglyned with maximized holding while imposed with minimizhear stress using the extruded quadrupole structure. Anrapping efficiency can be increased significantly using aetrical trapezoidal arrangement of the posts, because it l

he potential energy barrier of loading the trap without affeche trap strength. Chou et al.[45] demonstrated the trapping ofE.oli and its separation from blood cells by electrodeless dirophoretic trapping fabricated in an insulation substrate cosed of geometrical constrictions. Arai and co-workers[46]eveloped a system for separation of a single micro-orga


Fig. 6. Overview of the microfluidic device. (A) Schematic representation of operation. (B) Pseudocolored scanning electron micrograph (SEM) showing a singletrap consisting of four electroplated gold electrodes arranged trapezoidally along with the substrate interconnects. (C) SEM of a completed 1× 8 trap array. Scalebars: (B) 20�m and (C) 100�m. Reprinted with the permission from[44], © 2002 American Chemical Society.

fabricated devices with interdigitated electrodes by making useof differences in dielectric properties of cells[48]. Recently, aspecially shaped non-uniform electric field could be generatedby microfabricated planar microelectrode arrays on an insultingglass, and it could be employed to develop a DEP microfluidicsystem which could fractionate intact biological cells in suspen-sion into subpopulations[49]. Taking advantage of aligned topand bottom electrodes in channels, Seger et al.[50] employeddeflective dielectrophoretic barriers for accurately handling ofcells under controlled pressure-driven liquid flows.

DEP which uses an ac electric field to separate target cellsfrom cell samples in a flow stream has been demonstrated tobe quite effective and selective in concentrating, manipulating,and separating cells and bacteria with different sizes simultane-ously. In the meantime, DEP suffers from low survival rate ofbiological cells in electrical field and complicated instrumenta-tion. While, with the development of microfabrication methodand appearance of newly available electrode geometries for DEP,DEP manipulation in microfluidics for cell-based assay will havemore applications in microfluidics.

2.5. Other manipulation

Minerick et al.[51] delineated various fundamental drivingforces in microfluidic devices with parallel and orthogonal elec-t siono Ther nged theo driv-i d cow anso ris-t ndp lity ot gtho UV

exposure[52]. Ultrasound standing wave radiation force andlaminar flow have been employed for cell washing and mixingby a continuous field-flow fractionation (FFF) approach. Whenoperating at the flow rate of 10.2 ml/min, typical transit timethrough the sound field was less than 0.5 s, which was signifi-cantly shorter than that required by a centrifuge[53]. Yoshidaet al. [54] described a methodology for manipulation and nav-igation of selected single living cells to a desired location ona patterned bio-microelectromechanical system (bio-MEMS)chip that is applicable to rapid and precise formation of cellpattern.

3. Cell treatment

It is crucial to integrate cell treatment steps on chip to developmicrofluidic devices for cell analysis. In order to achieve analy-sis of cell constituents, cell lysis is absolutely necessary. Whencarrying out in vitro experiments, cell culture on chip is indis-pensable. Besides cell lysis and cell culture, achieving on chipcell electroporation and optoporation and cell fusion is also a hottopic in microfluidics. In the following section, the basic conceptof cell lysis, cell culture, cell electroporation and optoporation,and cell fusion devices are presented with selected examples.

3.1. Cell lysis

siso orta-b sp ssay.St ll idics f pro-t ears.A ys-t l not

rode configurations by examining the motion of a suspenf erythrocytes in response to a high-frequency ac field.esults show that electrode repulsion was so far the stroriving force in an ac field (far larger than DEP forces) andrthogonal electrode configurations could provide stronger

ng force than parallel electrode configurations. Soper anorkers[52] studied electrokinetic cell manipulation by mef electroosmotic flow in microfluidic devices fabricated in p

ine and UV-modified poly(methyl methacrylate) (PMMA) aolycarbonate (PC). Their results demonstrated the feasibi

ailoring the electroosmotic flow by changing the ionic strenf the buffer and modifying the polymer material through

st

-

f

The ability of integrating the lysis of cells with analyf their contents would greatly increase the power and pility of many microfluidic devices[1]. The choice of lysirotocol is important for the performance of subsequent aeveral methods of cell lysis, such as thermal lysis[55], elec-

rical lysis [56], mechanical lysis[57–60] as well as chemicaysis [61–64] have been successfully applied into microfluystems. Because thermal lysis may result in denaturation oeins[59], few researchers employed this method these ys almost all electrical lysis components in microfluidic s

ems are similar for cell-based assay, this lysis method wil


be discussed in this subsection. Chemical and mechanical lysisprinciples will be presented in detail with selected examplesbelow.

For mechanical-based lysis, nanostructured filter-like con-tractions in microfluidic channels with pressure-driven cell floware employed. Prinz et al.[57] utilized diffusive mixing to lyseE. coli cells and trapped the released chromosome by meansof DEP in an integrated microfluidic chip. Lysis of the cellswas achieved by rapid diffusional mixing of water with theosmotically stressed cells. Madou and co-workers[58] devel-oped a microfluidic compact disc (CD) platform using sphericalparticles for mechanical lysis of cells with a lysis efficiencyof approximately 65%. The rimming flow established inside apartially solid–liquid mixture-filled annular chamber when themicrofluidic CD rotated around a horizontal axis of rotationcould be employed for cell lysis. Lee and co-workers[59] fab-ricated nanoscale barbs in microfluidic devices for mechanicalcell lysis. It is shear and frictional forces induced by subject-ing cells to enter microfluidic filter region with sharp nanoscalebarbs that could be utilized to lyse cells. At a flow rate of300 ml/min within the filter region total protein and hemoglobinaccessibilities of 4.8 and 7.5% were observed, respectively,as compared to 1.9 and 3.2% for a filter without nanostruc-tured barbs. Taylor et al.[60] demonstrated cell disruptionsusing ultrasonic energy transmitted through a flexible inter-face into a liquid region and measured the released nucleica

entd s. Am l atn rapidc hro-c ievew e a5 viced dfl f 25-p thec ators[ ofz flu-i ieveb m)o on-c ande line(

protoc d lysc r toa erat d bef icesi lengT sis im te-g

3.2. Cell culture

Microfluidic devices are especially suitable for biologicalapplications, particularly on cellular level, because scale of chan-nels is commensurate with that of cells’[65], and scale ofthe devices allows important factors (e.g., growth factors) toaccumulate locally forming a stable microenvironment for cellculture[66]. Compared to traditional culture tools, microfluidicplatforms provide much greater control over cell microenviron-ment and rapid optimization of media composition using rela-tively small numbers of cells. Because a group of cells can moreeasily maintain a local microenvironment within microchan-nels than in macroscale culture flasks, cells grow significantlyslower in microchannels than cells in traditional culture flasks[67]. Microfluidic devices have been fabricated with three dif-ferent materials (silicon, PDMS, and borosilicate) and tested forembryo culture. Results showed that using microfluidic tech-nology for in vitro production of mammalian embryos has greatpotentials[68]. And the results from the perfusion culture offetal human hepatocytes (FHHs) in microfluidic bioreactorsaddressed the efficacy of such microstructures in future livertissue engineering[69]. In this subsection, selected examples ofmicrofluidic devices for cell culture will be reviewed.

PDMS has became one of the materials extensively used inmicrofluidic devices, due to its biocompatibility, low toxicity,high oxidative and thermal stability, optically transparent, lowp herm sings thea wereiT ur-f typed pesc rmalP uenceo ed dif-f nota ndt rolif-e ofe andc uresi da ellsi n ofm era izationo formmm romc art-m crossc to an asingo an-n of

cid.Chemical lysis uses non-ionic, less denaturing deterg

elivered from reservoirs or generated on chip for cell lysiicrofluidic lysis device that deals with erythrocyte removaearly the single-cell level was fabricated and evaluated forhemical lysis of erythrocytes. The complete lysis of erytytes and up to 100% recovery of leukocytes has been achithin 40 s[61]. Single-cell capture and chemical lysis insid0-pl closed volume was demonstrated in a microfluidic deesigned by Toner and co-workers[62]. In this device, cells anuids were independently isolated in two microchambers ol volumes using the geometry of the microchannels andoordinated action of four on-chip thermopneumatic actu62]. Beebe and co-workers[63] demonstrated the removalona pellucida from mammalian embryos in a PDMS microdic chip using chemical treatments. Zona removal was achy briefly washing with lysing agent (acid Tyrode’s mediuver the embryo. A microfluidic chip was developed forhip cell lysis based on local hydroxide electro-generationvaluated by lysis of red blood cells (RBCs), human tumorHeLa), and Chinese hamster ovary (CHO) cell lines[64].

Because extensive experiences and well-establishedols for large scale of samples are available, chemical-basean be easily integrated with microfluidics. However, in ordechieve efficient deliveries of lysis reagents or on-chip gen

ion of lysis reagents, more microfluidic components shoulabricated, that will increase the complexity of the micro-devn a way. Mechanical-based lysis also faces the same chalhough not discussed in this section, electrical-based lyost widely used in microfluidic cell lysis due to its easy inration and simple structure.

s

d

d

-is

-

e.s

ermeability to water, and low electrical conductivity, furtore, it could be easily fabricated into microstructures u

oft-lithography. The influence of PDMS composition onttachment and growth of several different types of cells

nvestigated thoroughly by Whitesides and co-workers[70].heir results showed that the ability of different PDMS s

aces for cell attachment, growth and proliferation was cell-ependent. No significant effect on cell growth for all cell tyould be observed when varying the composition of noDMS by excess base, and cells were able to reach confln these surfaces. PDMS having excess curing agent caus

erences in the growth of HUAEC and HeLa cells, but didffect the growth of 3T3 fibroblasts and MC3T3-E1 cells. A

he stiffness of substrate did not influence attachment and pration for all cell types[70]. Patterned multiple laminar flowstching solutions in capillaries was employed by Whitesideso-workers[71] to achieve easy adjusting topographical feat

n PDMS for mammalian cell culture. Hung et al.[72] presentehigh aspect ratio PDMS microfluidic device for culturing c

nside an array of microchambers with continuous perfusioedium. The high aspect ratio (∼20) between the microchambnd the perfusion channels offers advantages such as localf cells inside the microchamber as well as creation of a uniicroenvironment for cell growth. Taylor et al.[73] reported aicrodevice for direct growth of neurites and their isolation f

ell bodies. This PDMS microchip consisted of two compents separated by a physical barrier. Neurites could grow a

ompartments while fluidic isolation was maintained dueumber of micrometer-size grooves embedded in them. Bn the low diffusion properties of laminar flows in microchels, a PDMS microfluidic device consisting of an array


microinjectors and a base-flow channel could be employed toachieve localized drug application to cell cultures[74]. Jeon andco-workers[75] developed a plasma-based dry etching methodthat enabled patterned cell culture inside microfluidic devicesby allowing patterning, fluidic bonding, and sterilization to becarried out in one single step.

To achieve the goal of long-term cell culture and culturingcells up to higher density and larger numbers, continuous nutri-tion and oxygen supply, and waste removal through the culturemedium have to be ensured. Takayama and co-workers[76]described use of horizontally oriented mini reservoirs arraysas a gravity-driven pumping system to generate multiple fluidstreams inside microfluidic cell culture channels at a constantflow rate for prolonged periods. Marharbiz et al.[77] presenteda microfabricated electrolytic oxygen generator for high-densityminiature cell culture arrays. Long term (over 2 weeks) culturesof muscle cells spanning the whole process of differentiationfrom myoblasts to myotubes has been reported by Folch andco-workers[78]. Fig. 7 shows the schematics of the microflu-dic system. This microdevice could provide accurate controlof perfusion rates using a gravitational flow and biochemicalcomposition of the environment surrounding cells. Prokop etal. [79] developed NanoLiter BioReactors (NBRs) for long-

term culture and maintaining up to several hundred culturedmammalian cells in volumes three orders of magnitude smallerthan those in standard multi-well screening plates. Refresh-able Braille display-based microfluidic bioreactors, which weremore densely packed and not limited to linear and unidirec-tional perfusion, were developed for cell culture up to 3 weeksunder perfusion[80]. Yasuda and co-workers[81] developed atype of on-chip microcultivation chamber, which could directlymeasure the valve opening/closing by optical microscope, forlong-term cultivation of swimming cells. Fujii and co-workers[82] achieved large scale cells (up to 107 cells/cm3) culture ina microfluidic device with a multilayer bioreactor containingan oxygen supply system. Lee and co-workers[83] presenteda microfluidic cell culture array which could assay 100 dif-ferent cell-based experiments in parallel for long-term cellularmonitoring. Repeated cell growth/passage cycles, reagent intro-duction, and real-time optical analysis could all be achieved inthis microdevice.

In order to improve the efficiency of cell culture, someapproaches such as fabricating three-dimensional microstruc-tures [84–86] and attempting other biocompatible materials[87,88] are presented. Yasuda and co-workers developed amethod which made use of non-contact 3D photo-thermal etch-

FmitsC

ig. 7. Schematics and operation of the microfluidic device. (A) The short-teicrofluidic perfusion network (LT-�FN) is shown in black. (B) Schematics of the

n dark gray and Layer II is shown in light gray. (C and D) Two modes of opeo “2” to maintain a viable cell culture; “3–5” are inlets and “6” is the outlet for toluble factor gets “squeezed” by two side streams; arrows indicate the direchemistry.

rm microfluidic perfusion network (ST-�FN) is shown in gray and the long-termfragment (boxed in (A)) of the master mold showing two layers: Layer I is shown

ration of the device: (C) medium is gravity-fed through the LT-�FN from inlet “1”he ST-�FN; (D) hydrodynamic focusing in the device: a central stream carrying ation of the flows. Reprinted with the permission from[78], © 2005 Royal Society of


ing with a 1480 nm[84] or 1064 nm[85] infrared focused laserbeam to form shapes of agar microstructures for cultivatingcells. Desai and co-workers[86] fabricated a 3D heterogeneousmultilayer tissue-like structure inside microchannels for cell cul-ture. Single- and multi-stepwise microstructures were fabricatedin photosensitive biodegradable polymers poly-(�-caprolactone(CL)-dl-lactide (LA)) tetraacrylate for static cell culture of HepG2 cells and fetal human hepatocyte (FHH) cells[87,88].

Other approaches for microfluidic cell culture were also stud-ied. Cell cultivation was performed in a highly parallelizedmanner in fluid segments that were formed as droplets at a chan-nel junction. Organic and cell containing aqueous phase weremerged at channel junctions[89]. Multichannel microelectrodearray which could influence and record electrical cellular activ-ity was integrated into the microfluidics for cell culture[90]. Agradient-generating microfluidic platform for optimizing prolif-eration and differentiation of neural stem cells in culture wasdescribed by Jeon and co-workers[91].

3.3. Electroporation, electrofusion, and optoporation

Introduction of hydrophilic or membrane-impermeantmolecules into cells is crucial for various cell-based assays. Alarge number of methods have been proposed, but only elec-trooporation, electrofusion, and optoporation have been success-fully applied into microfluidic systems for cell treatment untiln pat-t akesm pli-cO h isq

oten-t heat,i ra-t hinfi thyl-m dt 293Tc ento ctro-p yedt midsa ropo-r eld,g veralf uc-i ells[ n-t ion.L atios lizedia tedl edc abi-l

Fig. 8. Schematic view of the device and protocol to fuse liposomes. Sev-eral electrode gaps have been designed. (1) Liposomes alignment (ac voltage).(2) Membrane breakdown (dc pulses). (3) Reconnection into a hybrid vesicle.Reprinted with the permission from[98], © 2004 Kluwer Academic Publishers.

Despite the efficiency that can be obtained by carrying outcell fusion within networks of microchannels, so far, few experi-ments have been performed to exploit its advantages. Tresset andTakeuchi[98] reported a microfluidic device with high aspectratio electrodes and low power consumption for electrofusion ofliposomes and cells with the fusion yield of 75%. A schematicview of the experimental protocol is depicted inFig. 8. Aftercells were introduced into the flow channel, a “pearl-chain”alignment of liposomes along the electric field lines could beobtained by applying an ac voltage. Then, a series of short andstrong dc pulses disrupted membranes in the region of contactof liposomes and initiated the fusion. Orwar and co-workers[99] developed a microfluidic device for combinatorial fusionof liposomes and cells. A large number of combinatorially syn-thesized liposomes with complex compositions and reactionsystems could be obtained from small sets of precursor lipo-somes.

Optoporation is a laser-based method in which the laserbeam interacts with an absorptive medium. When interactingwith nearby cells, a mechanical transient or stress wave, whichcan produce temporary alterations in the plasma membrane,can be produced because of optical breakdown, ablation, orrapid heating of the absorbing medium. Due to its complicatedoptical setup and complex operation, few microfluidic devicesintegrated with optoporation have been reported. Nevertheless,Allbritton and co-workers[100]had characterized cellular opto-p her uidicd

ieven oid-i heatm theya calfi t ofw ationa apy.

ow. The ease with which arrays of microelectrodes can beerned and integrated with networks of microchannels microfluidic systems an especially attractive platform for ap

ations in cell electroporation[92–97]and electrofusion[98,99].ptoporation offers advantage of remote operation whicuite desirable for some bio-applications[100].

Because electroporation on a microchip overcomes the pial risks of using a high voltage and generating excesst is widely used for in vitro gene transfection. Electropoion microchips consisting of microchannels with gold tlm electrodes on both sides were fabricated in polymeethacrylate (PMMA)[92] or on glass slide[93] and evaluate

hem by continuous gene transfection using Huh-7 andell lines. Lin et al.[94] studied the site-specific enhancemf in vitro gene delivery using electrostatic forces and eleoration microchips. An attracting electric field was emplo

o induce electrophoresis of negative charged DNA plasnd to concentrate them near cell surface prior to electation. Compared to that without the attracting electric fiene transfection efficiency could be enhanced up to se

olds. Electroporation microchips were developed for introdng genes not only into cell lines but also into primary c95]. Huang and Rubinsky[96] developed a microchip for corolled electroporation of single cells in a flow-through fashoaded cells were transported precisely to the electroporite by a microfluidic channel and were electropermeabindividually with virtually 100% gene transfer rate. They[97]lso developed an electroporation microchip that incorpora

ive biological cell in its electrical circuit and thereby inducontrolled electroporation in the cell for membrane permeization at a single-cell level.

n

a

oration in microfluidic systems using visible laser light. Tesults showed that optoporation is attainable on microflevices.

Electroporation and electrofusion microchips can achecessary electric field with a much lower applied voltage, av

ng the potential risks of using high voltage, and dissipatingore quickly due to its large surface/volume ratio. Thoughlso suffer from low survival rate of biological cells in electrields, their advantages have not been fully exploited. A loork should be done to take full advantages of electropornd electrofusion microchips to achieve on-chip gene ther


4. Cell analysis

Miniaturization of analytical devices promises numerousbenefits, including reduced cell consumption, automated andreproducible reagent delivery, and improved performance. Infollowing sections, microfluidic devices with specific bioana-lytical applications of cell processing are reviewed.

4.1. Micro cytometer

Cytometry refers to measurement of physical and chemicalcharacteristics of cells. Current flow cytometry is restricted to‘benchtop’ or full-sized laboratory models which are expensiveand generally dedicated to single use technology. Develop-ment of readily adaptable chip technology will allow the flowcytometer unit to be adaptable to a wide variety of potentialuses with portability and low cost. Recently, micro cytometershave become a vital component in any micro total analysis sys-tem aiming to investigate biological events at the single-celllevel [101]. Table 1lists some successful attempts toward microcytometer.

Differing from conventional flow cytometers which usuallyemploy fluorescence detection, microfluidic flow cytometry canbe mounted on a variety of detection instruments. Schwille andco-workers[105]developed a micro cytometer which supportedconfocal detection of fluorescent cells and particles and subse-q ct tos riseda ing,h tec-t r tod hiche outt d cow olwp uret eeni nceo anca alsohd entl icallyt

anyo picar rkers[ po-r lticp ls, top hionT rtings timec rugee ro-o nd a

high gain photodiode directly over the sorting microchannel intheir micro cytometer. Wolff et al.[112] developed a pressure-driven microfabricated high-throughput fluorescent activatedmicro cytometer which integrated the “smoking chimney” struc-ture for hydrodynamic focusing of sample and cell sorting, aholding and culturing chamber and waveguides for optical detec-tion of cells. The fluorescent activated cell sorting at a samplethroughput as high as 12,000 cells/s at 100-fold enrichment canbe achieved using this microchip. Several different optical ele-ments such as waveguides, lens and fiber-to-waveguide couplerswere incorporated into a microfluidic device designed by Wolffet al.[114] for sorting and analyzing different kinds of cells andparticles. Three different signals (forward scattering, large anglescattering, and extinction) were measured simultaneously. Sker-los and co-workers[118]developed a multi-angle micro cytome-ter in which integrated solid-state lasers and silicon-based PINphotodiodes were employed to perform flow cytometry mea-surements. The observation cell is designed with integratedmicrogrooves that permit the physical registration of opticalfibers which served as optical waveguides for laser excitationand fluorescence detection. The multi-angles design could helpto increase signal-to-noise ratio and to achieve multiple simulta-neous cytometric measurements at a single interrogation point.Nucleic acid-labeled fungus (S. cerevisiae) can be sorted at afast count rate of 500 particles/s using this microchip.

There are also some other approaches of microfluidic cells cs owc llsfl hip,t rtedb ers[ uidd etryt er-e theirs utm orl sisr justb twom to bea sedl rtingr p to2 results

4

entalb simul-t ithins sen-s n ofs flu-i

uently allowed for their sorting in the fluid phase with respepectroscopic properties. This micro flow cytometer compbranched channel system which could initiate fluid mix

ydrodynamically focus the sample solution for confocal deion and allow direct implementation of reaction steps prioetection and sorting. A deformability based cell sorter, wmployed optical stretcher for deformability detection with

he need for specific labeling, was developed by Guck anorkers[106]. Gawad et al.[109] developed a cytological tohich was based on the micro Coulter particle counter (�CPC)rinciple for cell counting and separation. The device meas

he spectral impedance of individual cells and allows scrng rates over 100 samples/s on a single-cell basis. Influef different cell properties, such as size, membrane capacitnd cytoplasm conductivity, on the impedance spectrumad been studied in detail[110]. Ramsey and co-workers[121]eveloped a micro flow cytometer which employed coincid

ight scattering and fluorescence to detect electrophoretransported and fluorescently labeledE. coli.

An integrated micro flow cytometer should incorporate mf the necessary components and functionalities of a tyoom-sized laboratory on to a small chip. Quake and co-wo101] developed a microfabricated cell sorter which incorated various microfluidic functionalities, including peristaumps, dampers, switch valves, and input and output welerform cell sorting in a coordinated and automated fashe whole cell sorting system is self-contained and the sochemes can be implemented on the device to performourse measurements on a single cell for kinetic studies. Kt al. [108] demonstrated the feasibility of integrating micptical components and using a flip-chip technique to bo

-

s-se,

l

.

-r

orter. Wang and co-workers[102] developed a microfluidiystem using gravity and electric force to drive cells for flytometry. Under driving of their own gravity force, ceowed passing the detection region in an upright microchen entered into the sorting electric field and were soy a switch-off activation program. Marr and co-work

113] exploited the inherent laminar nature of microscale flynamics and incorporated applied fields and image cytom

o develop micro cytometer. Any visually identifiable diffnces between cells or particles could be employed fororting. Recently, Wang et al.[115]described a high-throughpicrofluidic cell sorter with an all-optical control switch f

ive cells. As shown inFig. 9, cells pass first through an analyegion, then through the optical switching region situatedefore a Y-shaped junction, where the flow splits intoicrochannels. When a cell is detected and determinedtarget cell, the optical switch will be activated and a focu

aser spot deflects the cell to the target output channel. Souns of cell populations ranging from as few as 1000 cells u80,000 cells can be completed in less than an hour. Thehowed a major improvement in purity and yields.

.2. Chemical cytometry

Understanding molecular mechanisms of many fundamiological processes and serious health disorders requires

aneous analysis of a large number of chemical species wingle cells. Chemical cytometry refers to the use of highitivity analytical tools to characterize chemical compositioingle cells[122]. In this section, selected examples of micro

dic chemical cytometry will be reviewed.


Table 1Recently published examples of micro cytometer

Featured items Driving force Detection method Ref.

A cell sorter which is incorporated with variousmicrofluidic functionalities, includingperistaltic pumps, dampers, switch valves, toperform cell sorting is described

Pneumatically actuated pumps and valves Fluorescence [101]

A method based on gravity and electric forcedriving of cell for flow cytometry in anupright microfluidic chip system is described

Gravity focusing electric switching Confocal fluorescence [102]

A mechanism for cell sorting in a microfluidicdevice with combined electroosmotic andhydrodynamic pressure-driven flow isdescribed

Hydrodynamic pressure and electroosmotic Fluorescence [103]

A valveless switch for microparticle sorting withthe electrophoretic switch technique and thetwo phases laminar flow streams is presented

Hydrodynamic electrophoretic switching Fluorescence [104]

A microfluidic chip, which comprises abranched channel system to initiate fluidmixing and to hydrodynamically focus thesample solution down to a thin flow layer forconfocal detection, is described

Hydrodynamic focusing electrokinetic switching Confocal fluorescence [105]

A deformability based cell sorter which employsthe optical stretcher for deformabilitydetection without the need for specificlabeling is described

Hydrodynamic pressure Optical stretcher [106]

A confocal microfluidic particle sorter usingfluorescent photon burst detection is presented

Electroosmotic focusing and switching Confocal fluorescence [107]

A microfluidic device which integratesmicro-optical components and sorts singlecells by means of an off-chip valve switchingtechnique is described

Hydrodynamic focusing off-chip valve switching Fluorescence [108]

A microfluidic cytological tool which measuresthe spectral impedance of individual cells orparticles and allows screening rates over100 samples/s on a single-cell basis isdescribed

Hydrodynamic focusing Impedance spectroscopy [109,110]

A microfluidic flow cytometer, which employselectrokinetic forces for flow focusing andsample switching, and incorporates buriedoptical fibers for the on-line detection of cellsor particles, is described

Electrokinetic focusing and switching Fluorescence [111]

A microfluidic fluorescent activated cell sorterwith several integrated, functional structuresis described

Hydrodynamic focusing and switching Fluorescence [112]

A method, which exploits the inherent laminarnature of microscale fluid dynamics andincorporates applied fields and imagecytometry to enable sorting based upon anyvisually identifiable difference between cellsor particles, is described

Hydrodynamic focusing optical switching Fluorescence [113]

A microchip flow cytometer, in which severaldifferent optical elements (waveguides, lensand fiber-to-waveguide couplers) areintegrated with microfluidic channels, isdescribed

Hydrodynamic focusing Scattered light [114]

An all-optical control switch for live cells and itsimplementation in a high throughput,fluorescence-activated microfluidic cell sorteris described

Hydrodynamic focusing optical switching Fluorescence [115]

Multi-channel device for cell or particlecounting and sorting utilizing a digital imagedetection system is described

Hydrodynamic focusing electrokinetic switching Digital image detection [116]

A microfluidic system that can separate motilesperm from small samples basing on theability of motile sperm to cross streamlines ina laminar fluid stream is described

Hydrodynamic pressure Fluorescence [117]


Table 1 (Continued )

Featured items Driving force Detection method Ref.

A PDMS based multi-angle microfluidic flowcytometer which used solid-state lasers andsilicon-based PIN photodiodes to performflow cytometry measurement is described

Hydrodynamic focusing Fluorescence [118]

The feasibility of using auto-fluorescence (AF)for the detection of single living cells forlabel-free cell sorting in microfluidic systemsis demonstrated

Electroosmotic focusing and switching Confocal fluorescence [119]

A microchip flow cytometer which uses anair-liquid two-phase microfluidic system toproduce a focused high-speed liquid samplestream of particles and cells is described

Aerodynamic focusing Fluorescence [120]

A microfluidic device which employescoincident light scattering and fluorescence todetect electrophoretically transportedfluorescently labeledE. coli is described

Electrophoretic focusing Coincident light scatteringand fluorescence

[121]

Some attempts focused on detection of small moleculeswhich are related to cell metabolism. The high-throughput chem-ical analysis of single cells was demonstrated by McClainet al. [123] on an integrated microfluidic device. The aver-age analysis rate was 7–12 cells/min, which is 100–1000times faster than standard benchtop CE analysis of singlecells. Fang and co-workers[124] demonstrated the feasibil-ity of integrating the whole process of single-cell analysison a microfluidic. A throughput of 15 samples/h and a reten-tion time precision of 2.4% R.S.D. (n = 14) were obtained.

FR

The average separation efficiency for GSH in lysed cellswas 2.13× 106 ± 0.4× 106 plates/m. A microfluidic chemicalcytometry device was developed and evaluated for electrophero-grams of amino acids from individual Jurkat T-cells. The totaltime for the analysis of one cell is less than 1 h[125]. Lilgeand co-workers[126]presented a microfluidic device for single-cell CE in parallel microchannels. Calcein-labeled acute myloidleukemia (AML) cells were selected and transported by opticaltweezers. After electromechanical lysis, contents of individualcells were electrophoreticly separated and the labeled cellularcontents were recorded by LIF detection.

Others concentrated on detection of large proteins. Amicrofluidic system integrating continuous lysis of bacterialcells and fractionation and detection of a large intracellular pro-tein (�-galactosidase) has been demonstrated[127]. Wainrightet al.[128] designed microfluidic devices which were amenableto isotachophoresis (ITP)-zone electrophoresis (ZE) separationsand evaluated them by analyzing a cell surface protease (ADAM17) in live intact THP-1 cells in physiological buffers with detec-tion limits below 10 cells/assay. Tabuchi et al.[129] reporteda pressurization technique for microchip electrophoresis thatenabled 15 s separation of 12 samples of complex protein mix-tures extracted from Jurkat cells.

Most attention was paid on the genetic assays. Quake andco-workers[130] developed a microfluidic chip for automatednucleic acid purification from small numbers of bacterial orm do ramo d inF oft intot rym or-o etely( na m

ig. 9. Layout of the microfluidic sorting junction and the optical switch.eprinted with the permission from[115], © 2005 Nature Publishing Group.

t forf tc am-

ammalian cells. As shown inFig. 10f, this chip consistef an actuation layer and a fluidic layer. Schematic diagf one instance of the DNA isolation process is depicteig. 10a–e. Bacterial cells (red), buffer (green) for dilution

he cell sample and lysis buffer (yellow) are introducedhe microfluidic chip first (Fig. 10a), and then into the rotaixer (Fig. 10b). The three different liquids are mixed thughly and consequently bacterial cells are lysed complFig. 10c). The lysate is flushed over a DNA affinity columnd drained (Fig. 10d). Finally, purified DNA is recovered fro

he chip by introducing elution buffer and can be usedurther analysis or manipulation (Fig. 10e). A microchip thaonsists of microfluidic mixers, valves, pumps, channels, ch


Fig. 10. DNA purification chip and schematic diagram of one instance of the DNA isolation process (open valve, rectangle; closed valve,× in rectangle). (a) Bacterialcells (red), buffer (green) for dilution of the cell sample and lysis buffer (yellow) are introduced into the microfluidic chip. (b) The cell sample, dilution buffer andlysis buffer slugs are introduced into the rotary mixer. (c) The three different liquids are mixed thoroughly and consequently bacterial cells are lysed completely.(d) The lysate is flushed over a DNA affinity column and drained to the ‘waste port’. (e) Purified DNA is recovered from the chip by introducing elution buffer andcan be used for further analysis or manipulation. (f) Integrated bioprocessor chip with parallel architecture. This chip consists of the actuation layer and the fluidiclayer. The width of the fluid channels is 100�m and that of valve actuation channels is 200�m. The chip has 26 access holes, 1 waste hole and 54 valves within20 mm× 20 mm. Reprinted with the permission from[130], © 2005 Nature Publishing Group. (For interpretation of the references to color in this figure legend, thereader is referred to the web version of the article.)

bers, heaters, and DNA microarray sensors was presented byLiu et al. [131] and evaluated it by pathogenic bacteria detec-tion from approximately milliliters of whole blood samples andsingle-nucleotide polymorphism analysis directly from dilutedblood. The on-chip analysis started with the preparation processof a whole blood sample, which included magnetic bead-basedtarget cell capture, cell preconcentration and purification, andcell lysis, followed by PCR amplification and electrochemicalDNA microarray-based detection[131]. A microfluidic devicewhich integrated selection of individual cells, cell lysis, andelectrophoretic separation of apoptotic DNA fragments wasfabricated on a CD-like plastic disk for detection of doxorubicin-induced apoptosis in individual cardiomyocytes[132]. A semi-quantitative reverse transcription polymerase chain reaction(RT-PCR) procedure based on combination of competimer tech-nology with microchip electrophoresis was developed for totalRNA extraction from human colon carcinoma cell line CaCo-2. The result confirmed the role of cholesterol as a positiveinducer of specific factors on ABCA1 and ApoAI gene expres-

sions that was well estimable at low phytosterol concentrations[133].

4.3. Biochemical sensing

Because of its unique advantages such as controllable trans-port, immobilization, and manipulation of biological moleculesand cells, as well as controllable manipulation of reagents suchas separation, mixing, and dilution, microfluidic systems showgreat potential for analysis of intracellular parameters and todetect presence of cell metabolites, even on a single-cell level.Several examples of recently published biochemical sensing ofdifferent analytes from various cells in microfluidic devices arepresented inTable 2.

A 3D microfluidic network system, in which the sensor cellswere immobilized in agarose, was developed and employed forbioassay based on the expression of a bioluminescence reportergene. In this experiment,E. coli strains having a plasmid withfirefly luciferase gene (luc) under transcriptional control of


Table 2Recently published examples of biochemical sensing using microfluidic devices

Type of study Cell type Detection method Ref.

Sensoring mutagenicity usingE. coli strainswhich have a plasmid with fireflyluciferase gene (luc) under transcriptionalcontrol of mutagen-induced bacterial SOSresponse

E. coli Bioluminescence [28]

Monitoring the ATP dependent calciumuptake reaction upon treatment of aconcentration gradient of a test solution

HL-60 cells Confocal fluorescence [31]

Studying the microfluidic bioreactor basedon hydrogel-entrappedE. coli for cellviability, lysis, and intracellular enzymereactions

E. coli Fluorescence [134]

Studying the application of PEG hydrogelmicrostructures encapsulating mammaliancells in high-throughput drug screening orpathogen detection

Fibroblasts, hepatocytes, macrophage Fluorescence [135]

Monitoring extracellular acidification ratesof a large number of cells(1× 105–1× 106)

Mouse hepatocyte cell line, TABX2S Absorption spectra [136]

Monitoring the kinetics of intracellularfluorescein diacetate (FDA) metabolismand calcium mobilization at single-celllevel, as stimulated by glucose and pHchanges

Yeast cell Fluorescence [137,138]

Monitoring allergic response, stimulatedwith dinitrophenylated bovine serumalbumin (DNP-BSA) after incubation withanti-DNP IgE)

Rat basophilic leukemia cell line,RBL-2H3

Fluorescence [139]

Monitoring l-glutamate and hydrogenperoxide for studying the effects of tracelevels of an endocrine disruptor, tributyltin(TBT)

Nerve cells Amperometric [140]

Quantify amounts of adenosine triphosphate(ATP), released from erythrocytes that aremechanically deformed as these cellstraverse microbore channels in microchips

Endothelial cells Chemiluminescence [141]

Probing naphthalene metabolism using invitro model

L2 (rat lung Type II epithelial cells),H4IIE (rat hepatocytes), HepG2/C3A (human hepatocyte) celllines

Fluorescence [142]

Measuring the bradykinin stimulation ofindividual cells by observing changes inintracellular Ca2+ levels usingCa2+-sensitive fluorescent dye

PC12 cells Fluorescence [143]

Monitoring real-time production and releaseof glucose and ethanol from immobilizedcells using chemiluminescent (CL)enzyme-based flow-through microchip

Yeast cells (Saccharomycescerevisiae)

Chemiluminescence [144,145]

Monitoring insulin secretion using on-lineelectrophoresis immunoassay

Islets of Langerhans Fluorescence [146,147]

Assaying cell viability, measuringionophore-mediated intracellular Ca2+

flux, and measuring multistepreceptor-mediated Ca2+

Jurkat T-cells, U93719 cells Fluorescence [148]

Monitoring the NAD(P)H and [Ca2+]i

responses under the stimulation of glucose�-Cells of Langerhans Fluorescence [149]

Real-time monitoring cellular glucoseconsumption and lactate extrusion byNAD-linked enzymatic analysis

Mouse hepatocyte cell lines,TABX2S and TABX1A

Absorption spectra [150]

Monitoring of cytochromec distribution in aneuroblastoma-glioma hybrid cell

NG 108-15 neuroblastoma gliomahybrid cell

Scanning thermal lens microscopeimages

[151]

Monitoring the quantum release of dopaminefrom the cell under the stimulation ofnicotine

PC12 cells Amperometric [152]


Table 2 (Continued )

Type of study Cell type Detection method Ref.

Dynamic measurements of lactate duringcell permeabilization

Cardiac myocytes (healthy andanoxic cells)

Amperometric [153]

Monitoring extracellular acidification rates(with pH-sensitive field effect transistors,ISFETs), cellular oxygen consumptionrates (with amperometric electrodestructures) and cell morphologicalalterations (with impedimetric electrodestructures, IDES) on single chips

LS174T adenocarcinoma colorectalcell line

Potentiometric, amperometric,impedimetric

[154]

Monitoring extracellular potential caused bytransmembrane currents associated withspontaneously initiated intracellularcalcium waves

Cardiac myocyte Potentiometric [155]

Studying the intracellular conversion offluorescein diacetate (FDA) to fluoresceinand the degradation of an inhibitoryprotein, I�B, as involved in the NF-�Bsignaling pathway

Jurkat T-cells yeast cells Fluorescence [156]

Long-term monitoring the intracellularreporter gene activity

HFF11 cell line Chemiluminescence [157]

Monitoring of expression events in singlecells by profiling the activation of thetranscription factor NF-�B in cells inresponse to varying doses of theinflammatory cytokine TNF-�

HeLa S3 cells Fluorescence [158]

Monitoring the oxygenation cycle of the celland to investigate effects likephoto-induced chemistry caused by theillumination.

Red blood cell Raman spectroscopy [159]

Monitoring cellular ATP ofE. coli E. coli Bioluminescence [160]

mutagen-induced bacterial SOS response was used as sensorbacteria for mutagenicity[28]. A microfluidic device that inte-grated cell docking and concentration gradient functions wasdeveloped for monitoring calcium uptake reaction of HL-60cells by Yang et al.[31]. In this microfluidic chip, an analytesolution could be diluted to different gradients as a function ofdistance along the dam structure. Real-time monitoring of ATPdependent calcium uptake reaction of HL-60 cells upon treat-ment of a concentration gradient of a test solution was achieved[31]. Li and Peng[137,138]developed a microfluidic system fordetecting cells’ fluorescein diacetate (FDA) metabolism and cal-cium mobilization at single-cell level, as stimulated by glucoseand pH changes. In this work, 3D liquid flow control conceptwhich employed special liquid flow fields to manipulate andretain a single cell freely in the chip was also presented. Noforce stronger than its gravitational force was exerted on thecell, which could be balanced on different positions on an arc-sloping wall, thus minimizing any negative impact on the cell dueto strong liquid flows. Spence and co-workers[141] describeda series of studies designed to quantify amounts of adeno-sine triphosphate (ATP). A comparison of the amounts of ATPreleased from red blood cells (RBCs) mechanically deformedin microbore tubing (2.54± 0.15�M) versus a microchip(2.59± 0.32�M) suggested that microchannels in PDMSmicrochips may serve as biomimetic arterioles in vivo. Shulerand co-workers[142] described a microscale cell culture analog( lly

based pharmacokinetics (PBPK) model, for realistically andinexpensively studying the adsorption, distribution, metabolism,elimination, and potential toxicity (ADMET) of chemicals. Sucha microfabricated device consists of a fluidic network of chan-nels to mimic the circulatory system and chambers containingcultured mammalian cells representing key functions of ani-mal “organ” systems. This paper described the application ofa two-cell system, four-chamber�CCA (“lung”–“liver”–“othertissue”–“fat”) device for proof-of-concept study using naph-thalene as a model toxicant. Naphthalene was converted intoreactive metabolites in the “liver” compartment, which then cir-culated to the “lung” depleting glutathione (GSH) in lung cells.Peterman et al.[143] presented a microfluidic device whichcould repeatedly deliver chemical compounds to multiple loca-tions and evaluated it by measuring the bradykinin stimulationof individual PC12 cells with a Ca2+-sensitive fluorescent dye.Basic computational results with experimental verification ofboth fluid ejection and fluid withdrawal by imaging pH changesusing a fluorescent dye also were reported in their work. Wheeleret al.[148] developed a microfluidic device which can passivelyand gently separate a single cell from the bulk cell suspen-sion, and precisely deliver nanoliter volumes of reagents to thatcell. The cell viability assays, ionophore-mediated intracellularCa2+ flux measurements, and multistep receptor-mediated Ca2+

measurements were demonstrated in this microfluidic device.A microfluidic device which is capable of partially stimulatinga
�CCA), which was a physical replica of the physiologica n islet and allows observation of the NAD(P)H and [Ca2+]i


responses was developed to test degree of electrical couplingwithin the islet. Several results were obtained in this research:the �-cells of an islet were sufficiently coupled to synchro-nize the [Ca2+]i response within glucose-stimulated regions,but were not coupled to the extent that allowed a glucose-stimulated [Ca2+]i response to be transmitted into neighboringnon-stimulated cells; Tolbutamide, an antagonist of the ATP-sensitive K+ channel, allowed�-cells [Ca2+]i oscillations totravel further into the non-stimulated regions of the islet. Theextent of Ca2+ propagation across the islet depended on a del-icate interaction between the degree of coupling and the extentof ATP-sensitive K+-channel activation[149]. Kitamori and co-workers[151] combined a scanning thermal lens microscopedetector with a cell culture microchip to realize on-chip cel-lular biochemical analysis. Cytochromec (cyt) release frommitochondria to cytosol during the apoptosis process was suc-cessfully monitored and directly imaged with extremely highsensitivity (down to 10−20 mol) and a high spatial resolutionof ∼1�m, without any labeling materials. Brischwein et al.[154]reported a multi-parametric silicon sensor chip device withpotentiometric, amperometric, and impedimetric microsensorsto enable investigations of cellular microphysiological patterns.The chip layout is shown inFig. 11: five pH-sensitive field effecttransistors (pH-ISFETs) for measurements of extracellular acid-ification; one interdigitated electrode structure (IDES) for detec-tion of cell morphological changes; one amperometric sensor

Fsotw

structure for assessment of cellular oxygen consumption (pO2sensor). Extracellular acidification rates, cellular oxygen con-sumption rates, and cell morphological alterations were moni-tored on single chips simultaneously for up to several days[154].Li et al. [156] developed a microfluidic device, which consistedof special U-shaped microstructures with openings parallel to theliquid flow and weirs perpendicular to the flow. The versatility ofthis microfluidic system was demonstrated by studying two cel-lular processes at single-cell level: the intracellular conversionof FDA to fluorescein; and degradation of an inhibitory protein,I�B, as involved in the NF-�B signaling pathway. A Jurkat cellexpressed with I�B-EGFP was also employed to probe any pos-sible action of an herbal compound, isoliquiritigenin (IQ), onthe degradation of I�B-EGFP. Emneus and co-workers[157]developed a microfluidic system for assaying of the intracellu-lar reporter gene activity in real time and determining conditionswhich could minimize cell stress and hence unspecific expres-sion of the reporter gene. Liu et al.[160] developed a microflu-idic device which could perform on-chip cell lysis and sampleextraction to monitor cellular ATP ofE. coli using biolumines-cence (BL) detection. BL detection of ATP and ATP-conjugatedmetabolites was realized by using firefly luciferin–luciferase BLsystem. Under optimized conditions, ATP analysis could be real-ized within 30 s with a detection limit down to 0.20�M and adynamic linear range over two orders of magnitude. No inter-ferences from ADP, AMP, and cell lysis detergent Triton X-100w

4

ingm han-n pacei avep ntali ughc tionc hputa high-t ongw hievea mix-i ents[

rizedi m-p lasspa rms ah he ionc em-b time,

ig. 11. Layout of the sensor chip used for the presented measurements. The chize is 7 mm× 14 mm. Five pH-sensitive field effect transistors (pH-ISFETs),ne interdigitated electrode structure (IDES) and one amperometric sensor stru

ure are combined into a single chip and used for cellular recordings. Reprinteith the permission from[154], © 2003 Royal Society of Chemistry.

m wn inF em-b rdingep eral

ip

c-d

ere found within analysis standard deviation.

.4. Electrical characterization and ion channel studies

Cells control their electrical activity by opening and closolecular-sized pores in cellular membranes called ion cels. The ability to measure ion channel function in outer s

s thus very important but difficult. Patch-clamping cells hroven to be a powerful technique for elucidation of fundame

on channel protein biophysics and for drug discovery. Althoonventional patch-clamp methods provide high informaontents, they are laborious and suffer from low througnd high overall cost. Several approaches for achieving

hroughput electrophysiology are under development, amhich microchip-based patch-clamp systems uniquely achigher degree of miniaturization, faster perfusion and

ng, and lower reagent cost without losing information cont161–166].

Comparison of different patch-clamp setups is summan Fig. 12 [161]. Traditionally, patch-clamp recording is accolished with the help of a micromanipulator positioning gipette under a microscope[162,163]. As illustrated inFig. 12A,cell membrane patch is sucked into the glass pipette and foigh electrical resistance seal. Current that passes through thannels in either the membrane patch or the whole cell mrane is then recorded at different bias voltages. In the meanost chip-based devices use the planar geometry shoig. 12B, where the patch pore is etched in a horizontal mrane that separates the top cell compartment from the recolectrode compartment[164–166]. Lee and co-workers[161]resented a multiple patch-clamp array chip by utilizing lat


Fig. 12. Schematic view of patch-clamp setups. (A) Traditional patch-clampbased on a glass micropipette. (B) On-chip planar patch-clamp. (C) Top viewof the patch-clamp array device (optical microscope image) showing the centechannel and 14 radial patch channels. The small circle indicates one of the patcsites. Reproduced with the permission from[161], © 2004 American Institute ofPhysics. (D) SEM image of micronozzle (diameter 4�m) embedded in a largerhole. Reproduced with the permission from[165], © 2002 American Instituteof Physics.

cell trapping junctions which provided multiple cell addressingand manipulation sites for efficient electrophysiological mea-surements at a number of patch sites. As shown inFig. 12C,this device geometry not only minimizes capacitive couplingbetween the cell reservoir and the patch channel, but also allowsimultaneous optical and electrical measurements for studyinroles of ion channels with respect to their cellular functions.Orwar and co-workers[162] presented a microfluidics–patchclamp platform for performing high-throughput screening andrapid characterization of weak-affinity ion channel–ligand inter-actions. Different agents and buffer solutions could be pre-sented in programmable sequence to the biosensors, which couyield high screening rates. This microfluidic platform may havepotential applications in competitive assays, and dose–responcharacterizations from which agonist–antagonist affinity andefficacy as well as receptor blockage mechanisms. They[163]also studied the electrical properties and mechanical stabilityof the seal using a microfluidic chip generating laminar flow in

open volumes. The results showed that by applying an externalforce in the form of a fluid flow to a patch-clamped cell actingparallel to the cell-pipet axis, it was possible to increase patch-clamp recording times by over 100% and decrease the electricalnoise level by 40%. Folch and co-workers[164] designed andfabricated microfluidic devices which incorporated arrays ofnanoholes for patch-clamp electrophysiological recordings ofion channel activities and could realize the focal delivery ofbiochemical factors to cell surface with submicron-scale preci-sion. 3D silicon oxide micronozzles (Fig. 12D) integrated intoa microfluidic device for patch clamping has been developed byLehnert et al.[165]. A cell could be positioned on the nozzle bysuction through the hollow nozzle that extended to the backsideopening of the chip. A microanalysis system for multi-purposeelectrophysiological analyses has been presented by Han et al.[166]. This system has the capability to perform whole cell patchclamping, impedance spectroscopy, and general extracellularstimulation recording using integrated multi-electrode config-urations.

There are also some other approaches toward electrical char-acterization of cells. Microfluidic devices for the culture andelectrical characterization of epithelial cell layers have beendeveloped by Hediger et al.[167,168]. The main goal was toachieve both cell culture and impedimetric and potentiometriccharacterization on a single device. Huang et al.[169]developeda method that employed a microfabricated device for evaluationo esis-t weref luat-i e ani em-b

4

asedt indi-v top minea vings tlyp hipa

n-ni sultss tratet llarya kage.A ped,i ide am inl nsestc ltsw oneo n of

rh

sg

ld

se

f membrane electrical properties of single cells. Electrical rances of dead (membrane impaired) cells and live cellsound to be significantly different. This suggested that evang membrane resistances of individual cells could providnstant and quantitative measurement for determining cell mrane integrity and cell viability at single-cell level.

.5. Whole cell assay

Recent advances in microfluidic techniques have increhe potential of high-throughput biochemical assays onidual mammalian cells. Of particular interest is the abilityarallelize up-front assay protocols and still be able to exand treat every individual cell in the assay separately retrieingle-cell event information[170]. Several examples of recenublished whole cell assay for different cells in microfluidic cre presented inTable 3.

Shelby et al.[171] employed PDMS microfluidic chaels to characterize the complex behaviors ofP. falciparum-

nfected erythrocytes under capillary-like conditions. The rehowed that mature forms of infected erythrocytes conceno very high parasitemia at the mouth of a blocked capind fresh, pliable erythrocytes can squeeze through a blocmicrofluidic assay for bacterial chemotaxis was develo

n which a gradient of chemoeffectors was established insicrochannelvia diffusion between parallel streams of liquid

aminar flow. The random motility and chemotactic respoo l-aspartate,l-serine,l-leucine, and Ni2+ of WT cells andhemotactic-mutant strains ofE. coli were measured. The resuere amazing: WT cells dispersed better than cells lackingr both of major receptors Tar and Tsr; low concentratio


Table 3Recently published examples of whole cell assay using microfluidic devices

Type of study Cell type Ref.

Analyzing programmed cell death dynamics using apoptosischip and a step towards an integrated apoptosis chip forhigh-throughput drug screening on a single cellular level

HL-60 cells [170]

Characterizing the complex behaviors ofP. falciparum-infectederythrocytes under capillary-like conditions

P. falciparum-infected erythrocytes [171]

Studying bacterial chemotaxis E. coli, WT cells [172]Analyzing muscle cell contraction upon chemical stimuli Rat ventricular cardiomyocytes [173]Investigating a non-destructive, non-contact single-cell based

differential screening method which can study cells at singlecells level in arbitrary conditions

E. coli [174]

Studying chemotaxis of metastatic breast cancer cells in EGFgradients of well-defined profiles

MDA-MB-231 cells [175]

Investigating the influence of substrate-bound gradients onneuronal development

Rat hippocampal neurons [176]

Studying the immunoelectrophoresis of red blood cells Sheep erythrocytes [177]Studying relationships between cell shape and function and

probing the effect of flow on cells of different shapesEndothelial cells [178]

Parallel, quantitative analysis of stem cell proliferation anddifferentiation

Adult hippocampal progenitor cells (AHPs) [179]

Studying cell adhesion and cell mechanics by manipulating thegeometry and surface chemistry of the microdevices

WT NR6 cells [180]

Cytotoxicity test NCI H-23 cell [181]Detecting differences between normal skin and melanoma cell

lines and measuring consistent difference in a celldifferentiation model

CCD-1037, A375 and HL-60 cell [182]

Assessing the effects of an anticancer agent (doxorubicin) oncells

Hep G2 cell line [183]

Detecting the biological important cytokine tumor necrosisfactor alpha (TNF-�)

Dendritic cells [184]

Cell infection within a microfluidic device using virus gradients Spodoptera frugiperda (sf9) [185]

l-aspartate (3.2 nM at the inlet) that generated a significantmigration of RP437 WT cells;l-leucine acted as an attrac-tant sensed by Tar and as a repellent sensed by Tsr[172]. Liet al.[173] developed a microfluidic chip, which was integratedwith a thickness-shear mode (TSM) acoustic wave sensor, formuscle cell contraction analysis upon chemical stimuli. Jeonand co-workers[175] developed a microfluidic device to studychemotaxis of metastatic breast cancer cells, MDA-MB-231, inEGF gradients of well-defined profiles. The result suggested thatMDA-MB-231 cancer cell chemotaxis depended on the shape ofgradient profile as well as on the range of EGF concentrations.Dertinger et al.[176] fabricated substrate-bound gradients ofproteins with complex shapes using laminar flows in microchan-nels to systematically investigate influences of protein gradientson neuronal development. The results showed that axon specifi-cation was oriented in the direction of increasing surface densityof laminin. Taupin and co-workers[179] developed a microflu-idic platform that enabled parallel, quantitative analysis of stemcell proliferation and differentiation. This approach offered theability to observe large numbers of adhesion-dependent stemcells that may not survive without paracrine signaling support.Using this platform, one could further interrogate the responseof distinct stem cell subpopulations to micro-environmentalcues (mitogens, cell–cell interactions, and cell–extracellularmatrix interactions) that governed their behavior. A PDMS basedmicrofluidic device, which could dilute drugs with a buffer solu-

tion with serially increasing concentrations, was fabricated toperform cytotoxicity tests[181]. Zhang et al.[182] developedan integrated time of flight (TOF) optophoresis system to con-sistently detect significant differences between normal skin andmelanoma cell lines (CCD-1037 and A375, respectively), andmeasure consistent difference in a cell differentiation model(HL-60 cell line with DMSO treatment).

5. Concluding remarks

After several years’ of development, microfluidic deviceshave testified themselves to be powerful tools for cell-basedanalysis. Various cell manipulation methods can be incorpo-rated into microfluidic devices for cell-based analysis, suchas magnetic, optical, mechanical, and electrical manipulationmethods. Magnetic method is a clean, versatile, and non-invasivecell manipulation method. Because of extensive experience andwell-established protocols for large scale of samples, magneticmanipulation method can be easily integrated with microfluidicsfor cell-based assay. With the generation of new magnetic beadmaterials and modification methods for conjugating various lig-ands to magnetic beads, it is believed that the magnetic manipu-lation method will become more efficient. Due to the non-contactand contamination-free manipulation process, optical manip-ulation attracts great attentions. But their complicated opticalsetup, complex operation, and expensive instrumentation limit


their further applications in microfluidics. Size-dependent filter-based microfluidic devices exhibit numerous advantages, suchas high labeling efficiency, short detection time, and high repro-ducibility based on simple and robust experimental procedures.However, their poor selectivity limits their further applicationsin microfluidics for cell-based assays. Other constriction struc-tures also face the same challenge. Microgripper can preciselymanipulate single cell without any damage, but the complicatedfabrication procedures make it difficult to be applicable in com-mon biology and chemistry labs. Oppositely, dam and sandbagstructures can be easily fabricated using various methods and canalso efficiently immobilize cells with minimal stress on them.Dam and sandbag structures hold promises in microfluidic cellanalysis. Although the coating method which use cell adhesionto capture cells suffers from relatively low immobilization effi-ciency, its high selectivity and easily modifying microchannelinterior surface still make it a promising method for developingmicrofluidic devices for cell-based assay. DEP, which used anac electric field to separate target cells from cell samples in aflow stream, has been demonstrated to be quite effective andselective in concentrating, manipulating, and separating cellsand bacteria with different sizes simultaneously. In the mean-time, DEP suffers from low survival rate of biological cells inelectrical field and complicated instrumentation. However, withthe development of microfabrication method and appearance ofnewly available electrode geometries for DEP, DEP manipula-t eda thodi eapp dicsA eset

seda tionc ter,c cels emic hodc ievee n ofl abric s ina lengE idicc en-t urec rentm o bes rtiesP turec orpor andc rdet resh tiona lectrfi the

potential risks of using high voltage and dissipating heat morequickly due to large surface/volume ratio. Though they still suf-fer from the low survival rate, their advantages have not beenfully exploited. Complicated optical setup and complex opera-tion limit applications of optoporation microchips. Miniaturiza-tion of analytical devices promises numerous benefits, includingreduced cell consumption, automated and reproducible reagentdelivery, and improved performance. Micro cytometer is themost successful applications of microfluidic systems for cell-based assays. Various detection methods such as confocal detec-tion, optical stretcher, impedance spectra, and coincident lightscattering detection can be incorporated into micro cytometersfor different applications. Many of the necessary componentsand functionalities of a typical room-sized laboratory, such aspumps, dampers, switch valves, input and output wells, and soon, can be integrated into a small chip for cell sorting. Chemi-cal cytometer can also be successfully achieved in microfluidicplatform. A variety of small molecules, large proteins and DNAcan be simultaneous analyzed in chemical cytometer chip withina single run. Especially arresting is the possibility to treat andanalyze single living cells using microfluidic devices. A vari-ety of intracellular parameters and cell metabolites have beenanalyzed and detected using microfluidics, which also demon-strates the versatility of microfluidics. With the employment ofnanotechnology, single cell-based microfluidic devices for vari-ous sophisticated experiments will be the future direction of thisr

A

ongK

R

5.

74

74

ner,

siol.

ore-

04)

ces

85

hys.

ion will have more applications in microfluidics for cell-basssay. Cell manipulation using magnetic and electrical me

n microfluidics is most promising due to their simple and cheripheral equipments and easy integration with microfluilot of work should be done to take full advantages of th

wo methods.The successful applications of microfluidics for cell-ba

ssay include cell lysis chip, cell culture chip, electroporahip, electrofusion chip, optoporation chip, micro cytomehemical cytometry, biochemical sensing chip, and wholeensing chip. On-chip cell lysis can be realized using chal, mechanical, and electrical methods. All these lysis metan be easily integrated with microfluidics. In order to achfficient deliveries of lysis reagents or on-chip generatio

ysis reagents, more microfluidic components should be fated, that will increase the complexity of the micro-deviceway. Mechanical-based lysis also faces the same challectrical-based lysis is the most widely used in microfluell lysis due to its easy integration and simplified instrumation and will obtain more applications. On-chip cell cultan be successfully achieved in microfluidic platform. Diffeaterials (silicon, PDMS, and borosilicate) have proved t

uitable substrate for cell growth. Due to its inherent propeDMS becomes dominant material for fabricating cell culhip. Micropumps and oxygen generators have been incated into microfluidics to achieve long-term cell cultureulturing cells up to higher density and larger numbers. In oo improve the efficiency of cell culture, 3D microstructuave also been fabricated in cell culture chip. Electroporand electrofusion microchips can achieve the necessary eeld with a much lower required applied voltage, avoiding

s

.

l-s

-

e.

,

-

r

ic

esearch area.

cknowledgement

This work is supported by the Research Grants Council, Hong (CityU Project No. 9040983, CityU 102904).

eferences

[1] H. Andersson, A. van den Berg, Sens. Actuators B 92 (2003) 31[2] D. Erickson, D.Q. Li, Anal. Chim. Acta 507 (2004) 11.[3] D.A. Reyes, D. Iossifidis, P.A. Auroux, A. Manz, Anal. Chem.

(2002) 2623.[4] P.A. Auroux, D. Iossifidis, D.R. Reyes, A. Manz, Anal. Chem.

(2002) 2637.[5] T. Vilkner, D. Janasek, A. Manz, Anal. Chem. 76 (2004) 3373.[6] M. Ozkan, M. Wang, C. Ozkan, R. Flynn, A. Birkbeck, S. Ese

Biomed. Microdevices 5 (2003) 61.[7] S.K. Sia, G.M. Whitesides, Electrophoresis 24 (2003) 3563.[8] H.P. Tai, M.L. Shuler, Biotechnol. Prog. 19 (2003) 243.[9] D. Huh, W. Gu, Y. Kamotani, J.B. Grotberg, S. Takayama, Phy

Meas. 26 (2005) R73.[10] M. Toner, D. Irimia, Annu. Rev. Biomed. Eng. 7 (2005) 77.[11] M. Berger, J. Castelino, R. Huang, M. Shah, R.H. Austin, Electroph

sis 22 (2001) 3883.[12] H. Lee, A.M. Purdon, R.M. Westervelt, Appl. Phys. Lett. 85 (20

1063.[13] V.I. Furdui, D.J. Harrison, Lab Chip 4 (2004) 614.[14] P. Grodzinski, J. Yang, R.H. Liu, M.D. Ward, Biomed. Microdevi

5 (2003) 303.[15] D.W. Inglis, R. Riehn, R.H. Austin, J.C. Sturm, Appl. Phys. Lett.

(2004) 5093.[16] K.H. Han, A.B. Frazier, J. Appl. Phys. 96 (2004) 5797.[17] S. Umehara, Y. Wakamoto, I. Inoue, K. Yasuda, Biochem. Biop

Res. Commun. (305) (2003) 534.


[18] J. Enger, M. Goksor, K. Ramser, P. Hagberg, D. Hanstorp, Lab Chip4 (2004) 196.

[19] R.A. Flynn, A.L. Birkbeck, M. Gross, M. Ozkan, B. Shao, M.M. Wang,S.C. Esener, Sens. Actuators B 87 (2002) 239.

[20] A.L. Birkbeck, R.A. Flynn, M. Ozkan, D.Q. Song, M. Gross, S.C.Esener, Biomed. Microdevices 5 (2003) 47.

[21] P.Y. Chiou, A.T. Ohta, M.C. Wu, Nature 436 (2005) 370.[22] M.P. MacDonald, G.C. Spalding, K. Dholakia, Nature 426 (2003)

421.[23] L. Zhu, Q. Zhang, H.H. Feng, S. Ang, F.S. Chau, W.T. Liu, Lab Chip

4 (2004) 337.[24] H. Mohamed, L.D. McCurdy, D.H. Szarowski, S. Duva, J.N. Turner,

M. Caggana, IEEE Trans. Nanobiosci. 3 (2004) 251.[25] J. Moorthy, D.J. Beebe, Lab Chip 3 (2003) 62.[26] L.R. Huang, E.C. Cox, R.H. Austin, J.C. Sturm, Science 304 (2004)

987.[27] A. Khademhosseini, J. Yeh, S. Jon, G. Eng, K.Y. Suh, J.A. Burdick,

R. Langer, Lab Chip 4 (2004) 425.[28] H. Tani, K. Maehana, T. Kamidate, Anal. Chem. 76 (2004) 6693.[29] A. Revzin, R.G. Tompkins, M. Toner, Langmuir 19 (2003) 9855.[30] N. Chronis, L.P. Lee, J. Microelectromech. S 14 (2005) 857.[31] M.S. Yang, C.W. Li, J. Yang, Anal. Chem. 74 (2002) 3991.[32] J. Yang, C.W. Li, M.S. Yang, Lab Chip 4 (2004) 53.[33] C.W. Li, C.N. Cheung, J. Yang, C.H. Tzang, M.S. Yang, Analyst 128

(2003) 1137.[34] J. Lahann, M. Balcells, H. Lu, T. Rodon, K.F. Jensen, R. Lange, Anal.

Chem. 75 (2003) 2117.[35] B.J. Kirby, A.R. Wheeler, R.N. Zare, J.A. Fruetel, T.J. Shepodd, Lab

Chip 3 (2003) 5.[36] J.D. Cox, M.S. Curry, S.K. Skirboll, P.L. Gourley, D.Y. Sasaki, Bio-

materials 23 (2002) 929.04)

p 5

ne,

phys.

esis

al.

74

TAS

a,

hip 4

s 24

ki,

hip 4

R.S.

ino,

Chip

[58] J. Kim, S.H. Jang, G.Y. Jia, J.V. Zoval, N.A.D. Silvab, M.J. Madou,Lab Chip 4 (2004) 516.

[59] D.D. Carlo, K.H. Jeong, L.P. Lee, Lab Chip 3 (2003) 287.[60] M.T. Taylor, P. Belgrader, B.J. Furman, F. Pourahmadi, G.T.A. Kovacs,

M.A. Northrup, Anal. Chem. 73 (2001) 492.[61] P. Sethu, M. Anahtar, L.L. Moldawer, R.G. Tompkins, M. Toner, Anal.

Chem. 76 (2004) 6247.[62] D. Irimia, R.G. Tompkins, M. Toner, Anal. Chem. 76 (2004) 6137.[63] H.C. Zerigue, M.B. Wheeler, D.J. Beebe, Lab Chip 5 (2005) 108.[64] D.D. Carlo, C. Ionescu-Zanetti, Y. Zhang, P. Hung, L.P. Lee, Lab Chip

5 (2005) 171.[65] G.M. Walker, H.C. Zeringue, D.J. Beebe, Lab Chip 4 (2004) 91.[66] M.J. Mahoney, R.R. Chen, J. Tan, W.M. Saltzman, Biomaterials 26

(2005) 771.[67] G.M. Walker, M.S. Ozers, D.J. Beebe, Biomed. Microdevices 4 (2002)

161.[68] S. Raty, E.M. Walters, J. Davis, H. Zeringue, D.J. Beebe, S.L.

Rodriguez-Zas, M.B. Wheeler, Lab Chip 4 (2004) 186.[69] E. Leclerc, Y. Sakai, T. Fujii, Biochem. Eng. J. 20 (2004) 143.[70] J.N. Lee, X.Y. Jiang, D. Ryan, G.M. Whitesides, Langmuir 20 (2004)

11684.[71] S. Takayama, E. Ostuni, X.P. Qian, J.C. McDonald, X.Y. Jiang, P.

LeDuc, M.H. Wu, D.E. Ingber, G.M. Whitesides, Adv. Mater. 13(2001) 570.

[72] P.J. Hung, P.J. Lee, P. Sabounchi, N. Aghdam, R. Lin, L.P. Lee, LabChip 5 (2005) 44.

[73] A.M. Taylor, S.W. Rhee, C.H. Tu, D.H. Cribbs, C.W. Cotman, N.L.Jeon, Langmuir 19 (2003) 1551.

[74] P. Thiebaud, L. Lauer, W. Knoll, A. Offenhausser, Biosens. Bioelec-tron. 17 (2002) 87.

[75] S.W. Rhee, A.M. Taylor, C.H. Tu, D.H. Cribbs, C.W. Cotman, N.L.

ke,

we,

14.fel,

004)

hys.

oeng.

ip

K.

ujii,

ujii,

.M.

Lab

, E.S.

ors

03)

[37] S.K. Murthy, A. Sin, R.G. Tompkins, M. Toner, Langmuir 20 (2011649.

[38] A. Revzin, K. Sekine, A. Sin, R.G. Tompkins, M. Toner, Lab Chi(2005) 30.

[39] W.C. Chang, L.P. Lee, D. Liepmann, Lab Chip 5 (2005) 64.[40] T.H. Chen, D.A. Small, M.K. McDermott, W.E. Bentley, G.F. Pay

Biomacromolecules 4 (2003) 1558.[41] C.F. Chou, J.O. Tegenfeldt, O. Bakajin, S.S. Chan, E.C. Cox, Bio

J. 83 (2002) 2170.[42] M. Durr, J. Kentsch, T. Muller, T. Schnelle, M. Stelzle, Electrophor

24 (2003) 722.[43] Y. Huang, S. Joo, M. Duhon, M. Heller, B. Wallace, X. Xu, An

Chem. 74 (2002) 3362.[44] J. Voldman, M.L. Gray, M. Toner, M.A. Schmidt, Anal. Chem.

(2002) 3984.[45] C. Chou, M. Morgan, F. Zenhausern, C. Prinz, R. Austin, micro

(2002) 25.[46] F. Arai, A. Ichikawa, M. Ogawa, T. Fukuda, K. Horio, K. Itoigaw

Electrophoresis 22 (2001) 283.[47] L. Cui, D. Holmes, H. Morgan, Electrophoresis 22 (2001) 3893.[48] H. Li, R. Bashir, Sens. Actuators B 86 (2002) 215.[49] Y.L. Li, K.V.I.S. Kaler, Anal. Chim. Acta 507 (2004) 151.[50] U. Seger, S. Gawad, R. Johann, A. Bertsch, P. Renaud, Lab C

(2004) 148.[51] A.R. Minerick, R. Zhou, P. Takhistov, H.C. Chang, Electrophoresi

(2003) 3703.[52] M.A. Witek, S.Y. Wei, B. Vaidya, A.A. Adams, L. Zhu, W. Stryjews

R.L. McCarley, S.A. Soper, Lab Chip 4 (2004) 464.[53] J.J. Hawkes, R.W. Barber, D.R. Emersonb, W.T. Coakley, Lab C

(2004) 446.[54] M. Yoshida, K. Tohda, M. Gratzl, Anal. Chem. 75 (2003) 4686.[55] L.C. Waters, S.C. Jacobson, N. Kroutchinina, J. Khandurina,

Foote, J.M. Ramsey, Anal. Chem. 70 (1998) 158.[56] J. Cheng, E.L. Sheldon, L. Wu, A. Uribe, L.O. Gerrue, J. Carr

M.J. Heller, J.P. O’Connell, Nat. Biotechnol. 16 (1998) 541.[57] C. Prinz, J.O. Tegenfeldt, R.H. Austin, E.C. Cox, J.C. Sturm, Lab

2 (2002) 207.

Jeon, Lab Chip 5 (2005) 102.[76] X.Y. Zhu, L.Y. Chu, B. Chueh, M.W. Shen, B. Hazarika, N. Phad

S. Takayama, Analyst 129 (2004) 1026.[77] M.M. Maharbiz, W.J. Holtz, S. Sharifzadeh, J.D. Keasling, R.T. Ho

J. Microelectromech. Syst. 12 (2003) 590.[78] A. Tourovskaia, X. Figueroa-Masot, A. Folch, Lab Chip 5 (2005)[79] A. Prokop, Z. Prokop, D. Schaffer, E. Kozlov, J. Wikswo, D. Clif

F. Baudenbacher, Biomed. Microdevices 6 (2004) 325.[80] W. Gu, X.Y. Zhu, N. Futai, B.S. Cho, S. Takayama, PNAS 101 (2

15861.[81] K. Takahashi, K. Orita, K. Matsumura, K. Yasuda, Jpn. J. Appl. P

42 (2003) L1104.[82] E. Leclerc, Y. Sakai, T. Fujii, Biotechnol. Prog. 20 (2004) 750.[83] P.J. Hung, P.J. Lee, P. Sabounchi, R. Lin, L.P. Lee, Biotechnol. Bi

89 (2005) 1.[84] K. Kojima, H. Moriguchi, A. Hattori, T. Kaneko, K. Yasuda, Lab Ch

3 (2003) 292.[85] H. Moriguchi, Y. Wakamoto, Y. Sugio, K. Takahashi, I. Inoue,

Yasuda, Lab Chip 2 (2002) 125.[86] W. Tan, T.A. Desai, Biomaterials 25 (2004) 1355.[87] E. Leclerc, K.S. Furukawa, F. Miyata, Y. Sakai, T. Ushida, T. F

Biomaterials 25 (2004) 4683.[88] E. Leclerc, F. Miyata, K.S. Furukawa, T. Ushida, Y. Sakai, T. F

Mater. Sci. Eng. C 24 (2004) 349.[89] K. Martin, T. Henkel, V. Baier, A. Grodrian, T. Schon, M. Roth, J

Kohler, J. Metze, Lab Chip 3 (2003) 202.[90] T.M. Pearce, J.A. Wilson, S.G. Oakes, S.Y. Chiu, J.C. Williams,

Chip 5 (2005) 97.[91] B.G. Chung, L.A. Flanagan, S.W. Rhee, P.H. Schwartz, A.P. Lee

Monuki, N.L. Jeon, Lab Chip 5 (2005) 401.[92] Y.C. Lin, C.M. Jen, M.Y. Huang, C.Y. Wu, X.Z. Lin, Sens. Actuat

B 79 (2001) 137.[93] Y.C. Lin, M.Y. Huang, J. Micromech. Microeng. 11 (2001) 542.[94] Y.C. Lin, M. Li, C.C. Wu, Lab Chip 4 (2004) 104.[95] Y.C. Lin, M. Li, C.S. Fan, L.W. Wu, Sens. Actuators A 108 (20

12.[96] Y. Huang, B. Rubinsky, Sens. Actuators A 104 (2003) 205.


[97] Y. Huang, B. Rubinsky, Sens. Actuators A 89 (2001) 242.[98] G. Tresset, S. Takeuchi, Biomed. Microdevices 6 (2004) 213.[99] A. Stro1mberg, A. Karlsson, F. Ryttsen, M. Davidson, D.T. Chiu, O.

Orwar, Anal. Chem. 73 (2001) 126.[100] J.S. Soughayer, T. Krasieva, S.C. Jacobson, J.M. Ramsey, B.J.

Tromberg, N.L. Allbritton, Anal. Chem. 72 (2000) 1342.[101] A.Y. Fu, H.P. Chou, C. Spence, F.H. Arnold, S.R. Quake, Anal. Chem.

74 (2002) 2451.[102] B. Yao, G.A. Luo, X. Feng, W. Wang, L.X. Chen, Y.M. Wang, Lab

Chip 4 (2004) 603.[103] R. Johann, P. Renaud, Electrophoresis 25 (2004) 3720.[104] T. Takahashi, S. Ogata, M. Nishizawa, T. Matsue, Electrochem. Com-

mun. 5 (2003) 175.[105] P.S. Dittrich, P. Schwille, Anal. Chem. 75 (2003) 5767.[106] B. Lincoln, H.M. Erickson, S. Schinkinger, F. Wottawah, D. Mitchell,

S. Ulvick, C. Bilby, J. Guck, Cytom. Part A 59A (2004) 203.[107] B.H. Kunst, A. Schots, A.J.W.G. Visser, Rev. Sci. Instrum. 75 (2004)

2892.[108] J. Kruger, K. Singh, A. O’Neill, C. Jackson, A. Morrison, P. O’Brien,

J. Micromech. Microeng. 12 (2002) 486.[109] S. Gawad, L. Schild, P. Renaud, Lab Chip 1 (2001) 76.[110] S. Gawad, K. Cheung, U. Seger, A. Bertsch, P. Renaud, Lab Chip 4

(2004) 241.[111] L.M. Fu, R.J. Yang, C.H. Lin, Y.J. Pan, G.B. Lee, Anal. Acta Chim.

507 (2004) 163.[112] A. Wolff, I.R. Perch-Nielsen, U.D. Larsen, P. Friis, G. Goranovic, C.R.

Poulsen, J.P. Kutter, P. Telleman, Lab Chip 3 (2003) 22.[113] J. Oakey, J. Allely, D.W.M. Marr, Biotechnol. Prog. 18 (2002) 1439.[114] Z. Wang, J. El-Ali, M. Engelund, T. Gotsd, I.R. Perch-Nielsen, K.B.

Mogensen, D. Snakenborg, J.P. Kutter, A. Wolff, Lab Chip 4 (2004)372.

gen,.F.

005)

, S.

ens.

ilson

K.

. 73

.E.

04)

98.rris,

75

Nat.

nal.

lec-

003)

[136] H.A. Erxleben, M.K. Manion, D.M. Hockenbery, L. Scampavia, J.Ruzicka, Analyst 129 (2004) 205.

[137] X.Y. Peng, P.C.H. Li, Anal. Chem. 76 (2004) 5273.[138] X.Y. Peng, P.C.H. Li, Anal. Chem. 76 (2004) 5282.[139] Y. Matsubara, Y. Murakami, M. Kobayashi, Y. Morita, E. Tamiya,

Biosens. Bioelectron. 19 (2004) 741.[140] R. Kurita, K. Hayashi, K. Torimitsu, O. Niwa, Anal. Sci. 19 (2003)

1581.[141] A.K. Price, D.J. Fischer, R.S. Martin, D.M. Spence, Anal. Chem. 76

(2004) 4849.[142] K. Viravaidya, A. Sin, M.L. Shuler, Biotechnol. Prog. 20 (2004) 316.[143] M.C. Peterman, J. Noolandi, M.S. Blumenkranz, H.A. Fishman, PNAS

101 (2004) 9951.[144] R. Davidsson, F. Genin, M. Bengtsson, T. Laurell, J. Emneus, Lab

Chip 4 (2004) 481.[145] R. Davidsson, B. Johansson, V. Passoth, M. Bengtsson, T. Laurell, J.

Emneus, Lab Chip 4 (2004) 488.[146] J.G. Shackman, G.M. Dahlgren, J.L. Peters, R.T. Kennedy, Lab Chip

5 (2005) 56.[147] M.G. Roper, J.G. Shackman, G.M. Dahlgren, R.T. Kennedy, Anal.

Chem. 75 (2003) 4711.[148] A.R. Wheeler, W.R. Throndset, R.J. Whelan, A.M. Leach, R.N. Zare,

Y.H. Liao, K. Farrell, I.D. Manger, A. Daridon, Anal. Chem. 75 (2003)3581.

[149] J.V. Rocheleau, G.M. Walker, W.S. Head, O.P. McGuinness, D.W. Pis-ton, PNAS 101 (2004) 12899.

[150] C.M. Schulz, L. Scampavia, J. Ruzicka, Analyst 127 (2002) 1583.[151] E. Tamaki, K. Sato, M. Tokeshi, K. Sato, M. Aihara, T. Kitamori,

Anal. Chem. 74 (2002) 1560.[152] W.H. Huang, W. Cheng, Z. Zhang, D.W. Pang, Z.L. Wang, J.K. Cheng,

D.F. Cui, Anal. Chem. 76 (2004) 483.nal.

, B.

J.P.

174.C.004)

sh,

S.

hys.

718.hiu,

rna-lysts

81

Actu-

tron.

ctu-

son,

100

[115] M.M. Wang, E. Tu, D.E. Raymond, J.M. Yang, H.C. Zhang, N. HaB. Dees, E.M. Mercer, A.H. Forster, I. Kariv, P.J. Marchand, WButler, Nat. Biotechnol. 23 (2005) 83.

[116] G.B. Lee, C.H. Lin, S.C. Chang, J. Micromech. Microeng. 15 (2447.

[117] B.S. Cho, T.G. Schuster, X.Y. Zhu, D. Chang, G.D. SmithTakayama, Anal. Chem. 75 (2003) 1671.

[118] Y.C. Tung, M. Zhang, C.T. Lin, K. Kurabayashi, S.J. Skerlos, SActuators B 98 (2004) 356.

[119] J. Emmelkamp, F. Wolbers, H. Andersson, R.S. DaCosta, B.C. WI. Vermes, A. van den Berg, Electrophoresis 25 (2004) 3740.

[120] D. Huh, Y.C. Tung, H.H. Wei, J.B. Grotberg, S.J. Skerlos,Kurabayashi, S. Takayama, Biomed. Microdevices 4 (2002) 141.

[121] M. McClain, C. Culbertson, S. Jacobson, M. Ramsey, Anal. Chem(2001) 5334.

[122] N.J. Dovichi, S. Hu, Curr. Opin. Chem. Biol. 17 (2003) 603.[123] M.A. McClain, C.T. Culbertson, S.C. Jacobson, N.L. Allbritton, C

Sims, J.M. Ramsey, Anal. Chem. 75 (2003) 5646.[124] J. Gao, X.F. Yin, Z.L. Fang, Lab Chip 4 (2004) 47.[125] H.K. Wu, A. Wheeler, R.N. Zare, PNAS 101 (2004) 12809.[126] N.R. Munce, J.Z. Li, P.R. Herman, L. Lilge, Anal. Chem. 76 (20

4983.[127] E.A. Schilling, A.E. Kamholz, P. Yager, Anal. Chem. 74 (2002) 17[128] A. Wainright, S.J. Williams, G. Ciambrone, Q.F. Xue, J. Wei, D. Ha

J. Chromatogr. A 979 (2002) 69.[129] M. Tabuchi, Y. Kuramitsu, K. Nakamura, Y. Baba, Anal. Chem.

(2003) 3799.[130] J.W. Hong, V. Studer, G. Hang, W.F. Anderson, S.R. Quake,

Biotechnol. 22 (2004) 435.[131] R.H. Liu, J.N. Yang, R. Lenigk, J. Bonanno, P. Grodzinski, A

Chem. 76 (2004) 1824.[132] K. Kleparnık, M. Horky, Electrophoresis 24 (2003) 3778.[133] A. Cantafora, I. Blotta, R. Rivabene, L. Pisciotta, S. Bertolini, E

trophoresis 24 (2003) 3748.[134] J. Heo, K.J. Thomas, G.H. Seong, R.M. Crooks, Anal. Chem. 75 (2

22.[135] W.G. Koh, L.J. Itle, M.V. Pishko, Anal. Chem. 75 (2003) 5783.

,

[153] X. Cai, N. Klauke, A. Glidle, P. Cobbold, G. Smith, J. Cooper, AChem. 74 (2002) 908.

[154] M. Brischwein, E.R. Motrescu, E. Cabala, A.M. Otto, H. GrotheWolf, Lab Chip 3 (2003) 234.

[155] A.A. Werdich, E.A. Lima, B. Ivanov, I. Ges, M.E. Anderson,Wikswo, F.J. Baudenbacher, Lab Chip 4 (2004) 357.

[156] P.C.H. Li, L. de Camprieu, J. Cai, M. Sangar, Lab Chip 4 (2004)[157] R. Davidsson, A. Boketoft, J. Bristulf, K. Kotarsky, B. Olde,

Owman, M. Bengtsson, T. Laurell, J. Emneus, Anal. Chem. 76 (24715.

[158] D.M. Thompson, K.R. King, K.J. Wieder, M. Toner, M.L. YarmuA. Jayaraman, Anal. Chem. 76 (2004) 4098.

[159] K. Ramser, J. Enger, M. Goksor, D. Hanstorp, K. Logg, M. Kall, LabChip 5 (2005) 431.

[160] B.F. Liu, M. Ozaki, H. Hisamoto, Q. Luo, Y. Utsumi, T. Hattori,Terabe, Anal. Chem. 77 (2005) 573.

[161] J. Seo, C. Ionescu-Zanetti, J. Diamond, R. Lal, L.P. Lee, Appl. PLett. 84 (2004) 1973.

[162] J. Sinclair, J. Olofsson, J. Pihl, O. Orwar, Anal. Chem. 75 (2003) 6[163] J. Sinclair, J. Pihl, J. Olofsson, M. Karlsson, K. Jardemark, D.T. C

O. Orwar, Anal. Chem. 74 (2002) 6133.[164] T.F. Kosar, N.L. Stucky, C. Chen, K.J. Kim, A. Folch, Seventh Inte

tional Conference on Miniaturized Chemical and Biochemical AnaSystems, 2003, p. 25.

[165] T. Lehnert, M.A.M. Gijs, R. Netzer, U. Bischoff, Appl. Phys. Lett.(2002) 5063.

[166] A. Han, E. Moss, R. Rabbitt, B. Frazier, microTAS (2002) 805.[167] S. Hediger, J. Fontannaz, A. Sayah, W. Hunziker, M. Gijs, Sens.

ators B 63 (2000) 63.[168] S. Hediger, A. Sayah, J. Horisberger, M. Gijs, Biosens. Bioelec

16 (2001) 689.[169] Y. Huang, N. Sekhon, J. Borninski, N. Chen, B. Rubinsky, Sens. A

ators A 105 (2003) 31.[170] A. Valero, F. Merino, F. Wolbers, R. Luttge, I. Vermes, H. Anders

A. van den Berg, Lab Chip 5 (2005) 49.[171] J.P. Shelby, J. White, K. Ganesan, P.K. Rathod, D.T. Chiu, PNAS

(2003) 14618.


[172] H.B. Mao, P.S. Cremer, M.D. Manson, PNAS 100 (2003) 5449.[173] P.C.H. Li, W.J. Wang, M. Parameswaran, Analyst 128 (2003) 225.[174] Y. Wakamoto, S. Umehara, K. Matsumura, I. Inoue, K. Yasuda, Sens.

Actuators B 96 (2003) 693.[175] S.J. Wang, W. Saadi, F. Lin, C.M.C. Nguyen, N.L. Jeon, Exp. Cell

Res. 300 (2004) 180.[176] S.K.W. Dertinger, X.Y. Jiang, Z.Y. Li, V.N. Murthy, G.M. Whitesides,

PNAS 99 (2002) 12542.[177] T. Ichiki, T. Ujiie, S. Shinbashi, T. Okuda, Y. Horiike, Electrophoresis

23 (2002) 2029.[178] B.L. Gray, D.K. Lieu, S.D. Collins, R.L. Smith, A.I. Barakat, Biomed.

Microdevices 4 (2002) 9.[179] V.I. Chin, P. Taupin, S. Sanga, J. Scheel, F.H. Gage, S.N. Bhatia,

Biotechnol. Bioeng. 88 (2004) 399.

[180] H. Lu, L.Y. Koo, W.M. Wang, D.A. Lauffenburger, L.G. Griffith, K.F.Jensen, Anal. Chem. 76 (2004) 5257.

[181] H. Bang, S.H. Lim, Y.K. Lee, S. Chung, C. Chung, D.C. Han, J.K.Chang, J. Micromech. Microeng. 14 (2004) 1165.

[182] H.C. Zhang, E. Tu, N.D. Hagen, C.A. Schnabel, M.J. Paliotti, W.S.Hoo, P.M. Nguyen, J.R. Kohrumel, W.F. Butler, M. Chachisvillis, P.J.Marchand, Biomed. Microdevices 6 (2004) 11.

[183] Y. Tanaka, K. Sato, M. Yamato, T. Okano, T. Kitamori, Anal. Sci. 20(2004) 411.

[184] S. Cesaro-Tadic, G. Dernick, D. Juncker, G. Buurman, H. Krop-shofer, B. Michel, C. Fattinger, E. Delamarche, Lab Chip 4 (2004)563.

[185] G.M. Walker, M.S. Ozers, D.J. Beebe, Sens. Actuators B 98 (2004)347.

microfluidics technology for manipulation and analysis of biological

Documents