-
REVIEW
Microfabricated analytical systems for integrated
cancercytomics
Donald Wlodkowic & Jonathan M. Cooper
Received: 2 February 2010 /Revised: 29 March 2010 /Accepted: 3
April 2010 /Published online: 25 April 2010# Springer-Verlag
2010
Abstract Tracking and understanding cell-to-cell variabil-ity is
fundamental for systems biology, cytomics andcomputational
modelling that aids e.g. anti-cancer drugdiscovery. Limitations of
conventional cell-based techni-ques, such as flow cytometry and
single cell imaging,however, make the high-throughput dynamic
analysis oncellular and subcellular processes tedious and
exceedinglyexpensive. The development of microfluidic
lab-on-a-chiptechnologies is one of the most innovative and
cost-effective approaches towards integrated cytomics.
Lab-on-a-chip devices promise greatly reduced costs,
increasedsensitivity and ultrahigh throughput by
implementingparallel sample processing. The application of laminar
fluidflow under low Reynolds numbers provides an
attractiveanalytical avenue for the rapid delivery and exchange
ofreagents with exceptional accuracy. Under these conditions,the
fluid flow has no inertia, enabling the precise dosing ofdrugs,
both spatially and temporally. In addition, byconfining the
dimensions of the microfluidic structure, itis possible to
facilitate the precise sequential delivery ofdrugs and/or
functional probes into the cellular systems. Asonly low cell
numbers and operational reagent volumes arerequired,
high-throughput integrated cytomics on a singlecell level finally
appears within the reach of clinicaldiagnostics and drug screening
routines. Lab-on-a-chipmicrofluidic technologies therefore provide
new opportuni-ties for the development of content-rich
personalizedclinical diagnostics and cost-effective drug discovery.
It is
largely anticipated that advances in microfluidic technolo-gies
should aid in tailoring of investigational therapies andsupport the
current computational efforts in systemsbiology.
Keywords Cytomics . Cytometry .Microfluidics .
Lab-on-a-chip . Real-time cell assays . Cell sorting
Introduction
Cell populations represent an intrinsically heterogenic
andstochastic system with a high level of spatio-temporalcomplexity
[13]. Temporal cell-to-cell variability arisesfrom subtle
fluctuations in the concentrations of regulatoryproteins, protein
oscillations, position in the cell cycle andthe activation of
multiple compensatory and failsafemechanisms (e.g. apoptosis,
autophagy) (Fig. 1) [36].Within such heterogenic clusters, multiple
variables oftenact at the same time while interconnected
molecularpathways provide adaptive and compensatory outcomes(Fig.
1). Importantly, fluctuations at a single cell level oftenlead to
profound changes in the structure of particular cellpopulation
[710]. These phenomena are particularlyimportant in cancer research
where the regulation of cancercell death and survival involves
rapid switches betweenboth stochastic and binary signalling events.
The level ofcomplexity, with numerous variables acting at the
sametime, requires multiparametric and dynamic investigation
oflarge numbers of single cells. This feat is still
largelyinaccessible by using conventional bioanalytical and
diag-nostic approaches [710]. Figure 1 describes a
hypotheticalsituation where three cancer cells derived from
theseemingly homogeneous population respond differentiallyto a drug
stimulus in time. Whilst expression of an arbitrary
D. Wlodkowic (*) : J. M. Cooper (*)The Bioelectronics Research
Centre, University of Glasgow,Oakfield Avenue,Glasgow G12 8LT,
UKe-mail: [email protected]:
[email protected]
Anal Bioanal Chem (2010) 398:193209DOI
10.1007/s00216-010-3722-8
-
molecule that supports cell proliferation and blocks celldeath
rapidly decreases in one cell (denoted cell 1) anothercell responds
to the drug with a substantial delay (denotedcell 2). Yet another
cell (denoted cell 3) is characterized byonly brief oscillation in
the protein level that is followed bya quick recovery (Fig. 1). As
a result of stochasticmolecular responses within the population of
cells one cellgains a survival advantage while two others are
eliminatedby drug-induced reduction of protein level and/or
itsactivity (Fig. 1) [5, 79]. The surviving cell can subse-quently
divide and expand in the presence of the druggiving rise to a
resistant clone (Fig. 1). This hypotheticalexample demonstrates
that evolving subpopulation struc-ture can be defined by single
cell stochastic reactions.These mechanisms provide e.g. means for
repopulation andemergence of resistant clones that are the basis of
refractoryand relapsed cancers [1, 7, 1012]. Figure 1 illustrates
thus
that understanding the cell-to-cell variability is
fundamentalfor systems biology and particularly important in
processesof high spatio-temporal complexity such as response
ofcancer cells to therapy (Fig. 1) [69]. The latter is a
criticalevent that defines tumour growth rate and response to
anti-cancer therapy and has recently provided a new frameworkfor
the rationally designed and molecular anti-cancertherapeutics. Yet
only by obtaining the real-time insightsinto the drugcell
interactions can one create information-rich data sets,
significantly improving the in vitro validationof molecular drug
targets [13, 14]. The possibility ofcontinuously tracking
individual cells from the time ofencountering a stress signal,
through the decision-makingand execution phases, also provides
previously inaccessibleinformation of how complex biological
systems progressfrom e.g. life-maintaining to death-allowing steady
states[79, 15, 16]. The most promising are, in this respect,
the
Fig. 1 Reactions of cancer cellsto therapeutic compounds
arestochastic in nature and can leadto a variable therapeutic
out-come. a A hypothetical clonalcancer population,
respondingdifferentially to a drug stimulus,in time is presented.
Stochasticemergence of two different phe-notypes (red, green) in
time isseen on the upper panel. Notethat whilst expression of
anarbitrary molecule that supportscell proliferation and blocks
celldeath rapidly decreases in onecell (denoted cell 1, blue)
an-other cell responds to the drugwith a substantial delay
(denotedcell 2, red). Yet another cell(denoted cell 3, green) is
char-acterized by only brief oscilla-tion in the protein level that
isfollowed by a quick recovery ofthe protein level and/or
activityprofile (lower chart). Coloursrepresent three distinct
subpo-pulations of cells. b Repopula-tion initiated by stochastic
cellresponses to an anti-cancer drug.Clonal cell subpopulation gain
asurvival advantage whilst twoothers are eliminated by drug-induced
reduction of proteinlevel and/or its activity (as dis-cussed in
a)
194 D. Wlodkowic, J.M. Cooper
-
conceptual multidimensional assays where the combinationof gene
delivery technology (genomics), functional anddynamic live-cell
analysis (cytomics) and intracellularantibody staining of selected
proteins, also known asimmunocytochemistry (proteomics) can provide
innovative,multivariate assays for high-content data mining
andenhanced elucidation of cell signalling pathways [13,
14,1618].
Microfabricated systems for single cytomics
Recent studies in systems biology have shed new light onthe
underlying molecular mechanisms of cell-to-cell vari-ability in
cancer cell decision making [412]. Mostimportantly it appears that
not only genetic and epigeneticdifferences between cancer cells,
but also extremely subtlechanges in protein concentrations and the
intrinsic stochas-ticity of biochemical reactions within the cell
signallingpathways contribute to the observed cell-to-cell
variability[412]. As a result even in a theoretically genetically
andepigenetically identical population of cells (only sister
cellsare deemed to be genetically and epigenetically identical)the
responses to anti-cancer drugs will always be stochasticand
cellular decisions will be probabilistic in nature [412].Dynamic
switches between the stochastic and all-or-nothing events are
difficult to record by using conventional,end-point approaches [15,
16, 19]. The major drawback ofsuch binary analysis (e.g. western
blotting, ELISA, QT-PCR, fluorimetry, spectrophotometry) is that
they are basedon analysis of a total cell population that averages
theresults from every given cell [16, 2023]. Importantly theyonly
capture a snapshot of the cellular reaction which isinherently a
stochastic system [20, 21]. Flow cytometry(FCM) is one of the
conventional technologies that can beused to overcome a frequent
problem of traditional bulktechniques [2023]. The major advantages
of FCM includethe possibility of multiparameter measurements (i.e.
thecorrelation of different cellular events at a time), single
cellanalysis (the avoidance of bulk analysis) and rapid
analysistime (measuring thousand of cells per second) [2126].
Acommon drawback of conventional FCM methods is,however, that cells
are suspended in a laminar stream offluid that is rapidly discarded
[27, 28]. As a result norepetitive analysis of time-resolved events
is possible.Furthermore, as only integrated fluorescence is
collectedby photomultiplier tubes (PMTs) and no imaging is
used,this design suffers from the loss of both temporalinformation
and spatial (subcellular data) that enables thecharacterization of
many morphological features [15, 16,27, 28].
Measurement of inter- and intra-cellular complexity thatcan
uncover e.g. cell-to-cell variability in cancer cell
decision making requires an in-depth 4D investigation ofcell
populations at a single cell level [15, 16, 1921]. Thedevelopment
of reliable methodologies to track the behav-iour of single cells
within subtle cell subpopulations is alsoof paramount importance in
clinical diagnostics andpersonalized therapy. Inherent limitations
of traditionalflow cytometry have recently stimulated the fast
develop-ment of slide-based cytometers and in-flow
imagingcytometers (multispectral imaging cytometry) that
combineadvantages of both flow cytometry and fluorescence
imageanalysis (FIA) [2729]. These innovative technologiesemploy
cytometric principles rather than conventionalimage analysis to
collect high-content data on single cells[2729]. When combined with
slide-based and/or micro-plate scanning and increasing integration
with Nipkowspinning disk confocal modules, these systems
providereasonable throughput and high-resolution screening
capa-bilities at a subcellular level. Yet a common drawback of
allconventional high-throughput analysers is bulkiness, equip-ment
and maintenance costs [25, 26]. Moreover, the highpower consumption
and requirement for highly trainedservice personnel precludes their
widespread use [3032].Furthermore, a considerable number of cells
and reagentsare usually required for each conventional cell
analysis(typically above 104/mL) and processing of the samplesprior
to analysis is time consuming, involving severalcentrifugation
steps [21]. As the cost and time savings playan ever increasing
role in drug discovery and medicaldiagnostics, enabling strategies
that can reduce expendi-tures while increasing throughput and
content of informa-tion from a given sample attract a mounting
interest withinthe biopharmaceutical community [3336].
Transfer of traditional methods to a microfabricatedformat
provides a means to increase both the resolution ofanalysis and
sampling throughput while reducing the costsof a single assay
[3638]. During the last decade, a range ofmicroarray technologies
have been developed [3941].Technological foundations initially
developed for DNAmicroarrays have recently provided the starting
point forprotein, carbohydrate and tissue microarrays that are
slowlyemerging as useful tools in both clinical diagnostics anddrug
discovery pipelines [4145]. They offer miniaturiza-tion, low
reagent consumption, automation as well asqualitative and
quantitative approaches to analyse geneand protein expression on a
population level [44]. They do,however, suffer from a lack of
capabilities to monitor singleliving cells in real time and as such
represent a binarysystem that averages the results from every given
cell whilecapturing a snapshot of the intermittent cellular
reaction[15, 16, 19, 20]. As a result current experimental
evidenceof transient and intermittent physiological processes
islikely biased by the intrinsic time delay between
reagentaddition, cell reaction and the ensuing analysis. The
advent
Microfabricated analytical systems for integrated cancer
cytomics 195
-
Fig. 2 Microfabricated (lab-on-a-chip) bioanalytical systems for
cell-based assays allow for extensive integration of on-chip
componentsand high-level of miniaturization. a Integrated
microfluidic cell cultureand lysis on a chip. Chip with six
separate devices filled with a dye forchannel visualization (left
panel). A magnified image of one chambershowing the trapping region
structure which consists of an array offour cell traps separated by
spacers (right panel). Scale bars are1.3 mm. (Reproduced with
permission from The Royal Society ofChemistry (RSC) from ref.
[119].) b A high aspect ratio microfluidicdevice for
high-throughput mammalian cell culture. (Reproduced withpermission
from The Royal Society of Chemistry (RSC) from ref.[75].) c
Integrated and automated microfluidic cell culture system.
Insets show a close-up of two culture chambers, with the
multiplexerflush channel in between them (left), the input
multiplexer, with on-chip peristaltic pump, a waste output for
flushing the mixer, and thecell input line (right). (Reprinted with
permission from ref. [120].Copyright 2007 American Chemical
Society.) d A multi-step micro-fluidic device for studying cancer
metastasis. A layout of the devicewith 6 reservoirs, inlet ports
for cell seeding and Matrigel loadingports (left panel). Cell
migration, transmigration and cell invasion areathat comprises
10-mm-wide by 150-mm-long microfabricated gaps(right panel).
(Reproduced with permission from The Royal Societyof Chemistry
(RSC) from ref. [121])
196 D. Wlodkowic, J.M. Cooper
-
of microfluidic lab-on-a-chip technologies and their
inte-gration to design micro total analysis systems (TAS) isone of
the most innovative approaches towards integratedcytomics and
improved drug screening routines (Fig. 2)[3638]. Microfabricated
lab-on-a-chip devices promise
greatly reduced equipment costs, increased sensitivity
andthroughput by implementing massive experimental paralle-lization
while performing analysis on a single cell level(Fig. 2) [37, 46,
47]. Most importantly, as only low cellnumbers and operational
reagent volumes are required,
Fig. 3 Microfabricated cell arrays for extended cell culture and
high-throughput analysis at a single cell level. a Microfabricated
cell arrayfor correlative high-content analysis of individual
non-adherent cells.Scanning electron microscope (SEM) image of
Jurkat T cells insidethe microfabricated glass cell array. Scale
bar 20 mm (left panel).Device allows for lucid optical examination
of cells by utilizing e.g.wide-field transmitted light (middle
panel) and confocal microscopy(two Rhodamine 123-stained MOLT-4
cells are shown) (right panel).(Reproduced with permission from The
Royal Society of Chemistry(RSC) from ref. [122].) b Proprietary
microfabricated high-density cellarray. Microwells 650650 m were
fabricated by anodically bondingthe silicon grid wafer to a 500-m
borofloat glass substrate (leftpanel). Cell proliferation analysis
on a high-density cell array (right
panel). Long-term clone formation was started with a single
K-562cell FACS sorted to one well and cultured for up to 2
weeks.(Reproduced with permission from Picovitro AB, Stockholm,
Swe-den.) c Cell microarray platform fabricated by a PEG
(poly(ethyleneglycol) diacrylate) hydrogel patterning on glass. SEM
image ofMOLT-3 leukaemic cells confined in 15 mm15 mm PEG
wells.(Reproduced with permission from The Royal Society of
Chemistry(RSC) from ref. [123].) d CellTRAY, a novel micro-etched
live-cellscreening technology. Independently addressable regions of
glass orplastic microwells allow for a multiplexed and
time-resolvedexperimentation at a single cell level. Data courtesy
of Dr CathyOwen reproduced with permission from Nanopoint Inc.
(Honolulu,Hawaii , USA)
Microfabricated analytical systems for integrated cancer
cytomics 197
-
dynamic cytomics on rare cell populations finally appearswithin
investigational reach (Fig. 2) [37, 47, 48]. LOCsystems also
provide innovative ways to simultaneouslyanalyse large population
of single cells where, to uncoverthe stochastic basis of cellular
decision making, each cellhas to be isolated from others to
minimize the influence ofextrinsic factors such as cell-to-cell
contacts and paracrinesignalling. Many microfluidic systems can
also track singlecell responses multiparametically, whereby the
position ofevery cell is encoded and spatially maintained
overextended periods of time. In this respect,
microfluidicplatforms can fundamentally enhance the
mathematicaloncology and systems biology efforts and provide
newvistas for a new generation of rationally designed anti-
cancer drugs. They will be, therefore, a valuable tool for
theemerging field of systems oncology and provide new vistasto
validate mathematical models on patient-derived cells.To reflect on
this microfluidics has already been heralded asan emerging
technology with a multitude of applications inhigh-throughput drug
screening routines, content-rich per-sonalized clinical diagnostics
and diagnostics in resource-poor areas (Fig. 2) [3638, 4953].
Living cell microarrays
In the post-genomic era the functional assessment of
newlyidentified genes and validation of potential therapeutic
Fig. 4 Dielectrophoresis (DEP)-based cell arrays. a DEP dynam-ic
array cytometer that allows forautomated loading, observationand
arbitrary sorting of cells aftertheir optical examination.
SingleDEP traps consist of four elec-troplated gold electrodes
ar-ranged trapezoidally. (Reprintedwith permission from ref.
[70].Copyright 2002 AmericanChemical Society.) b Prototypeof a
negative dielectrophoresis(nDEP) trapping array for tran-sient
immobilization of singlecell. The ring-shaped nDEP trapswere
fabricated from two titani-um/platinum layers with a
ben-zocyclobutene (BCB) dielectric(left panel). c Bead and/or
cellimmobilization using in a nDEPis possible even in a
continuousflow of media (right panel).(Reproduced with
permissionfrom The Royal Society ofChemistry (RSC) from ref.
[68].)
198 D. Wlodkowic, J.M. Cooper
-
targets are a primary challenge [54, 55]. The recentintroduction
of molecularly targeted anti-cancer therapeu-tics has revealed a
significant need for new classes ofclinical bioassays [5659]. The
shift from cytotoxicchemotherapy to biomolecular therapeutics
requires thatdrug efficacy be assessed at the biologically active
dose thatmodulates the target, rather than through a
conventionaldosetoxicity relationship [5659]. In this context,
func-tional cytomics is slowly becoming an omnipotent part ofthe
post-genomic drug discovery pipelines [13, 14].
Multidimensional and high-throughput analysis of live cellscan
serve as an excellent dynamic analytical tool, forinstance to (i)
understand micro-evolution of cancer cells,(ii) discover function
of new genes and (iii) screen large-scale chemical and genomic
libraries [1316, 19, 20].Cellular processes such as cancer cell
responses to drugshave large cell-to-cell variations, and are often
initiated and/or executed in a multi-organelle/multi-pathway
fashion [16,19, 60, 61]. Moreover, the majority of events, e.g.
(i)mitochondrial or ER/Golgi dynamics during execution of
Fig. 5 Integrated microfluidicdevices for a long-term
cellculture. a A microfluidic devicefor a high-throughput
mamma-lian cell culture. Stable gradientgeneration can be created
acrossthe columns of independentmicrochamber arrays (left pan-el).
SEM image of singlemicrochamber (right panel).Multiple perfusion
channelssurround the main culturechamber that is 40 m in heightwith
a diameter of 1 mm.(Reproduced with permissionfrom The Royal
Society ofChemistry (RSC) from ref.[75].) b Microfluidic living
cellarray for real-time gene expres-sion studies. Multilayer
designconsists of a 1616 array of cellculture chambers that are
isolat-ed by sets of reversible PDMSbarriers. PDMS barriers
arecontrolled by valve controlmanifolds (left panel). Phasecontrast
image of a single cellculture chamber with H35 cellsreaching
confluence (right pan-el). (Reproduced with permis-sion from The
Royal Society ofChemistry (RSC) from ref.[76].) c Application of a
micro-fluidic living cell array (as de-scribed in b) for a
real-time geneexpression analysis usingtime-lapse microscopy and
ge-netically modified reporter cells.Fluorescence time-lapse
imagesof NFB and GRE GFP-reporters in microfluidic cellculture
chambers 2, 5, 8, 11, 14and 17 h after stimulation.(Reproduced with
permissionfrom The Royal Society ofChemistry (RSC) from
ref.[76].)
Microfabricated analytical systems for integrated cancer
cytomics 199
-
200 D. Wlodkowic, J.M. Cooper
-
caspase-dependent apoptosis, (ii) autophagy and
(iii)phagocytosis, all have highly ordered interactions
betweendifferent intracellular compartments [5, 8, 9, 60].
Monitor-ing of such biological phenomena at a single cell
levelrequires real-time and spatial detection systems with
drugdelivery resolution on a micrometre scale [15, 16, 19, 20].This
is currently not attainable in any macroscale analyticalsystems. As
a result many attempts have been made totransfer single cell assays
to a microfabricated systemcommonly known as a microarray format
[54, 55].Especially work by the group of Sabatini provided
newinsights on how to apply static high-density cell arrays
toperform discrete and parallel transfection of cells withthousands
of RNAi reagents [54, 55]. Adhering andgrowing cells reportedly
internalize the nucleic acids andbecome transfected (the remaining
cells form a non-transfected monolayer) [55]. Sabatinis group has
proventhat this reverse transfection array format is capable
ofefficient and spatially resolved cell transfection of humancells
[55]. Patterned RNAi cell microarray technology isundoubtedly a
major step towards the miniaturization andsimplification of
high-throughput cell assays in a micro-array format. Others
attempts have been made to use a
high-density living cell array platform where cells aregrown in
microwells with sizes of the order 10600 m(Fig. 3) [6267]. These
are fabricated in glass or biocom-patible polymers such as
poly(dimethylsiloxane) (PDMS)using standard microfabrication
techniques used in theelectronics industry (Fig. 3) [6267]. Many
differentdesigns have been proposed that support
large-scalesingle-cell trapping and real-time cell imaging using
micro-well arrays (Fig. 3) [6267]. This is a very
promisingtechnique mostly due to its simplicity of fabrication
andstraightforward operation [6267]. Trapping single cells insuch
static systems can be, however, influenced by manyfactors such as
cell size, cell buoyancy, microwell diameter,microwell depth and
settle time [6365]. Moreover theprecise delivery and exchange of
reagents using static cellmicroarrays still requires macroscale
liquid handling equip-ment [66, 67]. Recently another innovative
technique togenerate ordered arrays of cells has been proposed by
usingelectromagnetic (dielectrophoretic; DEP) pattering (Fig.
4)[6870]. To create spatially defined cell arrays the DEPtechnique
generally confines cells by exploiting theirelectric dipoles that
are induced in the electric field gradient[47, 48, 6870]. The cell
trapping element is usuallycomposed of a metal ring electrode and
the adjacent groundplane of the chip that creates a closed electric
field cage(Fig. 4) [6870]. DEP allows for a stable immobilization
offlowing cells even when they flow in a fluid stream ofrelatively
high velocity. Moreover, by controlling individ-ually addressable
electrodes, selected single cells or smallclusters of cells can be
released from the DEP trappingregion at any given time [6879]. This
opens vastopportunities for integrated devices with multiple
analyticalcapabilities on one chip [40]. DEP array technologies
are,however, complex and expensive to manufacture as theyrequire
fabrication of multiple electrode units for everysingle cell trap
[40, 6870]. Moreover, they require highconductivity physiological
media and generate high inten-sity electric field regions
(especially in positive DEP) thatsubject cells to large
transmembrane potential changes.New reports suggest, however, that
the negative DEP(nDEP) technology can overcome such limitations of
DEP(Fig. 4) [68]. nDEP trapping maintains levitating cellsinside
the potential energy wells [68].
All platforms described so far are based an open-accessformat.
As such they are prone to the substantial evapora-tive water losses
that hamper their exploitation in long-termand live-cell screening
experiments. Such a format alsoprecludes a straightforward strategy
for secure biocontain-ment of infectious specimens such as viral
gene vectors orHIV+ and blood samples. Microfluidics, however,
isuniquely aimed at manipulating liquids at ultralow volumesin
enclosed circuitry on-chip (Fig. 5) [37, 38, 4652]. Atmicroscale,
fluids exhibit different physico-chemical prop-
Fig. 6 Microfluidic cell arrays exploiting the hydrodynamic
celldocking principle and continuous microperfusion for drug
delivery. aDynamic single cell culture array with a branching
architecture andindividual chambers containing arrays of
micro-mechanical traps (leftpanel). A 3D diagram of the design and
mechanism of hydrodynamiccell trapping (middle panel). Traps are
fabricated in PDMS andbonded to a glass substrate (right panel).
They allow a gentle trappingwith no cell deformation. (Reproduced
with permission from TheRoyal Society of Chemistry (RSC) from ref.
[77].) b High-densitymicrofluidic array for cell cytotoxicity
analysis. Schematic of the 2424 chamber microfluidic cytotoxicity
array. Each chamber containseight micro cell sieves for cell
trapping (left panel). Microfluidiccytotoxicity array chip assembly
with fluidic interconnections (middlepanel). HeLa cell capture in
micro cell sieves (right panel). (Reproducedwith permission from
The Royal Society of Chemistry (RSC) from ref.[78].) c Microfluidic
array cytometer with a triangular chamber thatcontains a
low-density cell positioning array (left panel). SEM image ofa
trapping array fabricated in a biocompatible elastomer, PDMS
(middlepanel). Dynamic analysis of drug-induced cytotoxicity on a
microfluidiccell array (right panel). Note the stochastic nature of
the anti-cancer drugaction. Gradual increase in staining with
annexin V marks theexternalization of phosphatidylserine (PS)
residues characteristic of earlyapoptotic stages (red) whereas
gradual plasma membrane permeabilityto PI represents progressive
destabilization of plasma membranestructure in cells undergoing
apoptosis (yellow). d Quantification oftime-resolved analysis at a
single cell level. Human promyelocyticleukaemia cells were cultured
on a microfluidic array cytometer asdescribed in c. Four
representative cells were selected and their fluore-scence
following incorporation of PI assessed as a mean
fluorescenceintensity (MFI). Note the stochastic response to a
pan-kinase inhibitorstaurosporine with a profound variability
between cells in population.Black line represents a point of no
return where initial destabilization ofplasma membrane to PI is
irreversible and cell undergoes rapid celldeath
R
Microfabricated analytical systems for integrated cancer
cytomics 201
-
erties, dominated by viscous rather than inertial forces [46,47,
71]. Fluid flow under low Reynold numbers is thereforelaminar and
mass transport is dominated by diffusion [46,71]. As a result,
several parallel streams of reagents canflow next to each other
without convection. By adjustingthe fluid flow rate, the interface
between such adjacentstreams can be precisely controlled [72, 73].
In addition, asthe interial forces are minimal, when pumping of the
fluidceases so does the movement of the fluid, enabling theprecise
delivery of the reagent, in both time and space.These properties
uniquely position microfluidic lab-on-a-chip technology to culture
living cells for extended periodsof time and deliver cell-permeable
fluorescent probes, drugsand growth factors to defined subcellular
microdomains(Fig. 5) [7274]. The confining dimensions of the
micro-fluidic structures also facilitate precise positioning of
singlecells and sequential delivery of drugs and/or
functionalprobes while continuously monitoring the cell
microenvi-ronment and cell responses (Fig. 5) [37, 47, 48,
7274].
Accordingly, recent works have proven that microfluidicdevices
are useful for high-throughput drug screening andparticularly
suitable for 4D studies on rare subpopulations,such as cancer and
haematopoietic stem cells (Figs. 5 and6) [16, 19, 20, 47, 7578].
Several groups have recentlyproposed conceptual designs that enable
culture of cells on-chip (Fig. 5) [16, 19, 20, 47, 7578], some of
which allowfor a rapid hydrodynamic positioning of single cells
insidethe microfluidic chips (Fig. 6) [16, 19, 20, 77, 78]. At
thecentre of this cell microarray technology lays a set
ofmicro-mechanical traps, designed to passively
immobilizeindividual cells into a predefined pattern (Fig. 6) [16,
19,77, 78]. Living cell microarrays constructed from abiologically
compatible polymeric substratum allow quan-titative analysis of the
dynamic events at a single cellresolution (Fig. 6) [15, 16, 19, 77,
78]. Unlike flowcytometry, however, measurements are made at
multipletime points, and in contrast to conventional
time-lapsemicroscopy, image analysis is greatly simplified by
arrang-ing the cells in a spatially defined pattern and by
theirphysical separation (Fig. 6) [15, 16, 19, 77, 78]. Wepostulate
that the combination of microfluidic cell arrayswith integrated
on-chip gene delivery technology(genomics), functional and dynamic
live-cell analysis(cytomics) and intracellular antibody staining of
selectedproteins (proteomics) can provide innovative,
multivariateassays for high-content data mining and enhanced
elucida-tion of cell signalling pathways (Figs. 5 and 6) [15, 16,
19,63, 79]. Microfluidic cell array technologies are
thereforeemerging as a novel high-throughput experimental
tech-nique that makes it possible to correlate pre-stress
cellphenotypes and post-stress outcomes at a single cellresolution
[54, 55]. Like DNA microarrays, it is poised tobring breakthrough
discoveries in oncology, immunology
and neuroscience, particularly if used in conjunction withRNAi
library screens, gene reporter systems, dynamicfunctional bioassays
and single cell proteomics [54, 55,63, 76, 79].
Microflow chip-based cytometry
As discussed above, flow cytometry is a powerfulanalytical and
diagnostic tool that leverages the multipa-rameter and high-speed
measurements at the single celllevel [2129]. It suffers, however,
from a high cost,complex operation and limited portability.
Microfluidicsoffers here an innovative route to supersede these
dis-advantages through the development of innovative micro-flow
cytometers (FCM), micro fluorescently activated cellsorters (FACS)
and micro magnetically activated cellsorters (MACS) (Figs. 7 and 8)
[8087]. The majoradvantage of microflow cytometry chips is that
they samplea greatly reduced number of cells when compared
withconventional FCM [8093]. This is of particular valuewhen
studying e.g. rare patient-derived cells, and monitor-ing their
reactions to therapeutic compounds. Cells onplanar microfluidic
chips can be hydrodynamically focusedto generate a single file
which can be interrogatedsequentially by independent laser beams
(Fig. 7) [80, 81,8793]. Recent developments of nonplanar
microchipsopen many innovative opportunities to obtain
confinementand regulation of laminar streams of cells in two or
even inthree dimensions (Fig. 7) [80, 81, 8793]. The enclosednature
of microflow cytometers make them particularlysuitable for the
analysis of highly infective samples. Thismay be of particular
importance during e.g. production ofmammalian viral vectors;
monitoring of HIV infections,viral-induced cell death in pulmonary
diseases or monitor-ing of cancer-targeted adenoviral therapy.
Pressure-based microflow cytometers can be devel-oped with flow
rate controlled either by step motor-driven syringe pumps, positive
air pressure applied toinput reservoirs or vacuum applied onto
output reservoirs(Fig. 7) [80, 81, 8793]. Whilst, in conventional
flowcytometers the transfer rates through the flow chamber canbe as
high as 104106 cells/s, most microfluidic planarchips maintain are
much lower transfer rates of 10300 cells/s [8790]. This is
advantageous for preservationof live cells. The reduced transfer
rates can be, in turn,effectively compensated for by parallel
processing andsimultaneous analysis of multiple and parallel
streams ofcells on each chip [80, 81, 8793]. Some of the mostrecent
innovations in the FCM chip design also facilitatethe collection of
undiluted cells following the microcytometric analysis, a feature
not attainable in anyconventional systems [8890]. This allows for
many
202 D. Wlodkowic, J.M. Cooper
-
Fig. 7 Innovative microflow cytometers (FCM). a Prototype of
theelectrical microcytometer with 3D hydrofocusing. Experimental
setup:syringe pumps control sample and sheath flows through the
sensingregion, which is connected to the external circuit via
micro-manipulators.(Reproduced with permission from The Royal
Society of Chemistry(RSC) from ref. [91].) b Prototype of the
lab-on-a-chip cytometer withintegrated microfluidic dye laser,
optical waveguides, microfluidicnetwork and photodiodes.
(Reproduced with permission from TheRoyal Society of Chemistry
(RSC) from ref. [108].) c Single-layerplanar microcytometer
utilizing innovative drifting based 3D hydrody-namic focusing.
(Reproduced with permission from The Royal Societyof Chemistry
(RSC) from ref. [92].) d Prototype of the microfluidic flowcell
with a two-stage cascaded hydrodynamic focusing, integrated
mirrors (green), grooves for optical fibres (yellow) and
fluidicconnections (blue) to externally mounted electrodes.
(Reproduced withpermission from The Royal Society of Chemistry
(RSC) from ref. [93].)e CellLab Chip (Agilent Technologies, Santa
Clara, CA, USA). Cross-sectional view of the microfluidic chip with
an optical layout of theAgilent 2100 Bioanalyzer (Agilent)
providing an external interface tothe chip-based cytometer.
Excitation light sources (LED and solid-statelaser) are depicted
together with light paths and corresponding detectors(FL1 and FL2).
Note that substantial fluorescence signal separation (FL1em 525 nm
vs. FL2 em 685 nm) alleviates need for a spectralcompensation.
Disposable microfluidic cartridge and microfluidicnetwork that
allows for a 2D hydrodynamic focusing of cells into asingle file
(dotted panel on right)
Microfabricated analytical systems for integrated cancer
cytomics 203
-
functional studies to be performed post-analysis on thegiven
cell sample.
Apart from the traditional fluorescence detection, spec-tral
impendence using the Coulter principle has also beenadapted for
on-chip cytometers to study the function of cellsize, cytoplasmic
resistance and membrane capacitance[9496]. Precise differential
white blood cell counts havealready been demonstrated by using the
on-chip Coulter
principle [97, 98]. Recent reports suggest, however, thateven
more high-throughput data can be obtained by usingin-flow
dielectric spectroscopy on-chip [99, 100]. In thisregards
innovative high-throughput screening (HTS) tech-nologies that are
developed in a miniaturized formatinclude the capacitance and
impedance cytometry [100102]. Moreover, a number of unconventional
cytometrictechnologies have recently been proposed for a non-
204 D. Wlodkowic, J.M. Cooper
-
invasive and real-time cell analysis on microfluidic chips.These
include real-time studies on a single cell level such
astime-of-flight (TOF) optophoresis and scanning thermallens
microscopy (TLM) [103107].
Current progress in on-chip cytometry leverages manyrecent
advances in microfluidic technology for the singlecell analysis
with the ultimate outcome to produce user-friendly, reasonably
priced and portable devices capable ofmultiparameter fluorescent
interrogation of single cells insuspension (Fig. 7). The ultimate
challenges for micro-fabricated cytometers remain in their robust
coupling withthe laboratory macroenvironment (Fig. 7) [108].
Microfabricated cell sorters
The accurate detection, quantification and separation ofsingle
cell clones is of paramount importance in clinicaldiagnostics, drug
discovery pipelines and ultimately inpatient-tailored therapies
[2326]. In this respect, flowcytometry still remains the technology
of choice, especiallyfor rapid quantification and cell separation
using high-speedfluorescently activated cell sorting (FACS) [2326].
Indeed,
modern electrostatic sorters support sorting algorithmsbased on
up to 16 optical parameters from a single cell,with acquisition
rates that exceed 25,000 events per second[2326]. This has opened
new horizons for cell biology,immunology and cancer research
[2326]. The widespreaduse of FACS is, nevertheless, severely
limited because of itshigh complexity, power consumption and
resulting intrinsiccost of the equipment and need for specialized
training.This restricts such equipment to only centralized
corefacilities [87]. As large reagent expenditures and
multiplesample processing steps are also required, it
profoundlydecreases efficiency and makes separation of
clinicallyrelevant cell subpopulations particularly challenging.
More-over, the high-pressure and electrostatic charge applied
tocells during FACS can adversely affect recovery of fragilecells,
such as apoptotic cells and cancer stem cells [87].
Not surprisingly, there is an increasing interest anddemand for
cost-effective and portable cell sorting systemsthat will
supplement conventional FACS especially in (i)high-throughput cell
separation during drug screeningroutines, (ii) clinical grade cell
sorting, (iii) diagnostic inresource-poor areas, (iv) military
operations and (v) humanspace exploratory missions [37, 8186,
109118]. In thiscontext microfluidics has an immense potential to
meetthese demands, due to the inherent ease of rapid prototyp-ing,
potential of flexible and scalable designs, enhancedanalytical
performance and economical fabrication (Fig. 8)[37, 8186, 109118].
Recently, development of manyinnovative microfluidic cell sorters
has been reported(Fig. 8). These include (i) fluorescently
activated cellsorters (FACS), (ii) in-flow micro magnetic cell
sorters(MACS) capable of rapidly deflecting
paramagneticallylabelled cells in a continuous stream of isotonic
buffer, (iii)integrated optofluidics microsystems for
Raman-activatedcell sorting (RACS) and (iv) functionalized
micropostarray sorter [8186, 109118].
Immunomagnetic cell separation on-chip is
particularlyadvantageous mainly due to its simplicity. It
typicallyemploys monoclonal antibodies conjugated with
super-paramagnetic particles used to separate cell subpopulationof
interest [86, 110115, 117]. Advantages of this techniqueover FACS
include minimal power consumption, substan-tial portability and
simplicity of operation. Great simplifi-cation of these laboratory
procedures will promote thefurther development of the MACS
technology [86, 110115, 117]. Furthermore, by employing enclosed,
disposablechip sorting cartridges, these designs will enable
clinicalgrade, sterile sorting without undesired aerosol
formationusually associated with conventional FACS [86,
110115,117]. Such design characteristics enable a secure sorting
ofhighly infectious specimens without containment cabinetsas
opposed to the dedicated rooms necessary for biohazardFACS sorting.
Ultimately, we also envisage implementation
R Fig. 8 Innovative microfluidic cell sorters. a Schematics of
theintegrated microfabricated cytometer and high-throughput
fluores-cently activated cell sorter. (Reproduced with permission
from TheRoyal Society of Chemistry (RSC) from ref. [82].) b SEM
image ofthe micro cell sorter chip with integrated
holding/culturing chamber asdescribed in a: a sheathing buffer
inlet, b chimney sample inlet, cdetection zone, d holding/culturing
chamber, e sieve to allow diffusionof nutrients and confinement of
cells, f channel for draining excessliquid during sorting and for
feeding fresh media to the cells duringcultivation. (Reproduced
with permission from The Royal Society ofChemistry (RSC) from ref.
[82].) c Integrated optofluidic Raman-activated cell sorter (RACS)
that combines multichannel microfluidicdevices with laser tweezers
Raman spectroscopy (LTRS) for delivery,identification and sorting
of individual cells. (Reproduced withpermission from The Royal
Society of Chemistry (RSC) from ref.[116].) d Principles of the
continuous flow magnetic separation on-chip. An inhomogeneous
magnetic field is applied perpendicular tothe direction of flow.
Magnetic particles or magnetically labelled cellsare attracted into
the field and thus deflected from the direction oflaminar flow.
(Reproduced with permission from The Royal Societyof Chemistry
(RSC) from ref. [113].) e Microfluidic chip forcontinuous sorting
by using free-flow magnetophoresis. (Reproducedwith permission from
The Royal Society of Chemistry (RSC) fromref. [115].) f Prototype
of the micromagnetic-microfluidic bloodcleansing device (MMBCD).
Transverse merging of microfluidicchannels allow for the rotation
of inlet fluid streams about thelongitudinal axis of the separation
channel. The MMBCD devicegenerates magnetic field gradients across
vertically stacked channelsto enable continuous and high-throughput
separation of fungi fromflowing whole blood. The device was
successfully used to cleanse80% of living fungal pathogens from
human whole blood flowing at arate of 20 mL/h. (Reproduced with
permission from The RoyalSociety of Chemistry (RSC) from ref.
[117].)
Microfabricated analytical systems for integrated cancer
cytomics 205
-
of low-cost optofluidics modules that would delivercomplementary
on-chip flow cytometric analysis. This canbe particularly valuable
e.g. for CD4+ lymphocytes count-ing and isolation in HIV/AIDS
disease monitoring in sub-Saharan Africa. Successful applications
of micro magneticcell sorting have already been shown in a gentle
separationof human lymphocytes, fibroblasts and apoptosing
cancercells [86, 110115, 117]. Considering the simplicity of
theon-chip cell sorting protocols, these platforms have a
widepotential to be used for automated diagnostic and
laboratoryroutines [37].
Although micro magnetic cell sorting on-chip is muchslower than
any currently available FACS sorters, innova-tive lab-on-a-chip
designs can provide substantial improve-ments. By leveraging e.g.
potential for modular designincreased throughput can be achieved by
a parallel sampleprocessing paradigm. For this purpose one can
takeadvantage of a well-known precedent in microprocessordesign and
implement a multi-core sorting module config-uration. We are
convinced that micro magnetic cell sortingtechnology will be
especially valuable in challengingenvironments, such as biomedical
research in developingcountries, field exploratory missions and
possibly evenspace medicine and exobiology.
Future outlook
DNA and protein microarray technologies are rapidlyevolving
fields that paved the way to modern clinicaldiagnostics, predictive
toxicology and molecular pharma-cology [4045]. Therapeutic targets
revealed by genetic andproteomic screens have to be, however,
thoroughly validat-ed by functional live-cell assays that resolve
the spatial andtemporal interrelationships in molecular signalling
net-works at a large scale [13, 14]. Yet it is still
particularlychallenging to quantify rapidly changing, i.e.
dynamic,phenomena, which are by definition impractical to study
byusing conventional binary approaches [16, 19, 20]. Dis-secting
such signalling complexity at the single cell levelcan be only
obtained by microsystems, whereby theposition of every cell is
registered and maintained overextended periods of time [16, 19, 37,
48]. In this context,microfluidic lab-on-a-chip technologies
provide uncompli-cated and effective solutions for low-cost and
high-throughput screening routines at a single cell level
[37,4649].
As discussed above, the microfluidic cell arrays providenew
opportunities for multivariate single cell analysis atreasonably
high data acquisition speeds [15, 16, 19, 20].They are, thus,
particularly attractive for the clinical anddiagnostic laboratories
as they allow rapid analysis of onlysmall amounts of
patient-derived cells [15, 16, 19, 20].
Most importantly, they provide sensitivity that often cannotbe
easily achieved with any conventional analytical plat-forms. As
such they can be applied in a number of areasincluding accelerated
anti-cancer drug discovery andtherapy, particularly in
high-throughput and high-contentdrug screening routines [3638, 54].
The key challengesstill, however, lie ahead and include on-chip
integration andsimplification of many functional components such
asexcitation and collection optics, fluidics, electronics andtheir
robust incorporation with the clinical and screeninglaboratories
infrastructure [46]. Recent advances in the fieldhave recently
provided new solutions such as computer-controlled microvalve
arrays, drug mixers and dispensersworking at sub-nanolitre volumes
[46]. Miniaturization ofother analytical components, such as
organic light emittingdiodes (OLEDs) and detectors, is also poised
to revolution-ize future design of clinical chip-based diagnostic
devices[46, 93]. We believe that progress in microfluidic
solutionswill provide new milestones for the advancement ofbenchtop
and user friendly flow cytometers and integratedcell sorters [5052,
80, 82, 88]. Lab-on-a-chip has alreadybeen realized as an emerging
technology with a multitudeof applications in high-throughput drug
screening routines,content-rich personalized clinical diagnostics
and improvedanalytical capabilities for resource-poor areas.
Furtherprogress in this field will lead to stand-alone
portabledevices that warrant accelerated drug discovery,
andultimately personalized therapeutic regimens.
Acknowledgments Supported by Biotechnology and
BiologicalSciences Research Council (BBSRC); Engineering and
PhysicalSciences Research Council (EPSRC) and the Scottish
FundingCouncil, under RASOR Program (Radical Solutions for
Researchingthe Proteome).
Authors thank Dr Sara Lindstrm from Picovitro AB
(Stockholm,Sweden) and Dr Cathy Owen from Nanopoint Inc. (Honolulu,
Hawaii,USA) for providing exemplary data on proprietary
microfabricatedanalytical systems. The authors declare no
conflicting financialinterest.
References
1. Foo J, Michor F (2010) Evolution of resistance to
anti-cancertherapy during general dosing schedules. J Theor Biol
263(2):179188
2. Perkins TJ, Swain PS (2009) Strategies for cellular
decision-making. Mol Syst Biol 5:326
3. Roach KL, King KR, Uygun BE, Kohane IS, Yarmush ML,Toner M
(2009) Biotechnol Prog 25(6):17721779
4. Perc M, Rupnik M, Gosak M, Marhl M (2009) Prevalence
ofstochasticity in experimentally observed responses of
pancreaticacinar cells to acetylcholine. Chaos 19(3):037113
5. Hawkins ED, Markham JF, McGuinness LP, Hodgkin PD (2009)A
single-cell pedigree analysis of alternative stochastic lympho-cyte
fates. Proc Natl Acad Sci U S A 106(32):1345713462
206 D. Wlodkowic, J.M. Cooper
-
6. Fakir H, Tan WY, Hlatky L, Hahnfeldt P, Sachs RK
(2009)Stochastic population dynamic effects for lung cancer
progres-sion. Radiat Res 172(3):383393
7. Enderling H, Anderson AR, Chaplain MA, Beheshti A, HlatkyL,
Hahnfeldt P (2009) Paradoxical dependencies of tumordormancy and
progression on basic cell kinetics. Cancer Res69(22):88148821
8. Lavrik IN, Eils R, Fricker N, Pforr C, Krammer PH
(2009)Understanding apoptosis by systems biology approaches.
MolBiosyst 5(10):11051111
9. Raychaudhuri S, Skommer J, Henty K, Birch N, Brittain T
(2010)Neuroglobin protects nerve cells from apoptosis by inhibiting
theintrinsic pathway of cell death. Apoptosis 15(4):401411
10. Milas L, Hittelman WN (2009) Cancer stem cells and
tumorresponse to therapy: current problems and future
prospects.Semin Radiat Oncol 19(2):96105
11. Davis AJ, Tannock JF (2000) Repopulation of tumour
cellsbetween cycles of chemotherapy: a neglected factor.
LancetOncol 1:8693
12. Stavreva N, Stavrev P, Fallone BG (2009) Probability
dynamicsof a repopulating tumor in case of fractionated
externalradiotherapy. Phys Med 25(4):181191
13. Trnok A, Bocsi J, Brockhoff G (2006) Cytomics-importance
ofmultimodal analysis of cell function and proliferation
inoncology. Cell Prolif 39(6):495505
14. Trnok A, Valet GK, Emmrich F (2006) Systems biology
andclinical cytomics: the 10th Leipziger Workshop and the
3rdInternational Workshop on Slide-Based Cytometry,
Leipzig,Germany, April 2005. Cytometry A 69(1):3640
15. Faley S, Seale K, Hughey J, Schaffer DK, VanCompernolle
S,McKinney B, Baudenbacher F, Unutmaz D, Wikswo JP
(2008)Microfluidic platform for real-time signaling analysis of
multiplesingle T cells in parallel. Lab Chip 8(10):17001712
16. Wlodkowic D, Faley S, Zagnoni M, Wikswo JP, Cooper JM(2009)
Microfluidic single-cell array cytometry for the analysisof tumor
apoptosis. Anal Chem 81(13):55175523
17. Yang Y, Adelstein SJ, Kassis AI (2009) Target discovery
fromdata mining approaches. Drug Discov Today 14(34):147154
18. Kim YS, Maruvada P, Milner JA (2008) Metabolomics
inbiomarker discovery: future uses for cancer prevention.
FutureOncol 4(1):93102
19. Faley SL, Copland M, Wlodkowic D, Kolch W, Seale KT,Wikswo
JP, Cooper JM (2009) Microfluidic single cell arrays tointerrogate
signalling dynamics of individual, patient-derivedhematopoietic
stem cells. Lab Chip 9(18):26592664
20. Wlodkowic D, Skommer J, McGuinness D, Faley S, Kolch
W,Darzynkiewicz Z, Cooper JM (2009) Chip-based dynamic real-time
quantification of drug-induced cytotoxicity in human tumorcells.
Anal Chem 81(16):69526959
21. Darzynkiewicz Z, JuanG, Li X, GorczycaW,Murakami T,
TraganosF (1997) Cytometry in cell necrobiology: analysis of
apoptosis andaccidental cell death (necrosis). Cytometry
27(1):120
22. Darzynkiewicz Z, Bedner E, Traganos F (2001) Difficulties
andpitfalls in analysis of apoptosis. Methods Cell Biol
63:527546
23. Melamed MR (2001) A brief history of flow cytometry
andsorting. Methods Cell Biol 63:317
24. Robinson JP (2006) Multispectral cytometry: the next
genera-tion. Biophotonics Int 10:3640
25. Bonetta L (2005) Flow cytometry smaller and better.
NatMethods 2:785795
26. Shapiro H (2004) The evolution of cytometers. Cytometry
A58:1317
27. George TC, Basiji DA, Hall BE, Lynch DH, Ortyn WE, PerryDJ,
Seo MJ, Zimmerman CA, Morrissey PJ (2004) Distinguish-ing modes of
cell death using the ImageStream multispectralimaging flow
cytometer. Cytometry A 59(2):237245
28. Zuba-Surma EK, Kucia M, Abdel-Latif A, Lillard JW
Jr,Ratajczak MZ (2007) The imagestream system: a key step to anew
era in imaging. Folia Histochem Cytobiol 45(4):279290
29. Kamentsky LA, Kamentsky LD (1991)
Microscope-basedmultiparameter laser scanning cytometer yielding
data compara-ble to flow cytometry data. Cytometry 12:381387
30. Janossy G, Shapiro H (2008) Simplified cytometry for
routinemonitoring of infectious diseases. Cytometry B Clin Cytom
74(Suppl 1):S6S10
31. Manian BS (2008) Affordable diagnosticschanging the
para-digm in India. Cytometry B Clin Cytom 74(Suppl 1):S117S122
32. Baum LL, Crowe S, Landay AL (2007) Advances in CD4
cellenumeration in resource-poor countries. Curr Opin HIV AIDS
2(3):234240
33. Mayr LM, Bojanic D (2009) Novel trends in
high-throughputscreening. Curr Opin Pharmacol 9(5):580588
34. Schmid EF, Ashkenazy R, Merson J, Smith DA (2009)
Willbiomedical innovation change the future of healthcare?
DrugDiscov Today 14(2122):10371044
35. Zaenker KS, Entschladen F (2009) Paving roads for new
drugsin oncology. Recent Pat Anticancer Drug Discov 4(2):137145
36. Manz A, Dittrich PS (2006) Lab-on-a-chip: microfluidics in
drugdiscovery. Nat Rev Drug Discov 5:210218
37. Hong J, Edel JB, deMello AJ (2009) Micro-and
nanofluidicsystems for high-throughput biological screening. Drug
DiscovToday 14(34):134146
38. Kang L, Chung BG, Langer R, Khademhosseini A
(2008)Microfluidics for drug discovery and development: from
targetselection to product lifecycle management. Drug Discov
Today13(12):113
39. Deeken J (2009) The Affymetrix DMET platform and
pharmaco-genetics in drug development. Curr Opin Mol Ther
11(3):260268
40. Gant TW, Zhang SD, Taylor EL (2009) Novel genomic methodsfor
drug discovery and mechanism-based toxicological assess-ment. Curr
Opin Drug Discov Devel 12(1):7280
41. Gomase VS, Tagore S, Kale KV (2008) Microarray: an
approachfor current drug targets. Curr Drug Metab 9(3):221231
42. Camp RL, Neumeister V, Rimm DL (2008) A decade of
tissuemicroarrays: progress in the discovery and validation of
cancerbiomarkers. J Clin Oncol 26(34):56305637
43. Ma H, Horiuchi KY (2006) Chemical microarray: a new tool
fordrug screening and discovery. Drug Discov Today
11(1314):661668
44. Sobek J, Bartscherer K, Jacob A, Hoheisel JD, Angenendt
P(2006) Microarray technology as a universal tool for
high-throughput analysis of biological systems. Comb Chem
HighThroughput Screen 9(5):365380
45. Uttamchandani M, Yao SQ (2008) Peptide microarrays:
nextgeneration biochips for detection, diagnostics and
high-throughput screening. Curr Pharm Des 14(24):24282438
46. Whitesides GM (2006) The origins and the future of
micro-fluidics. Nature 442:368372
47. El-Ali J, Sorger PK, Jensen KF (2006) Cells on chips.
Nature442:403411
48. Sims CE, Allbritton NL (2007) Analysis of single
mammaliancells on-chip. Lab Chip 7:423440
49. Wang J, Ren L, Li L, Liu W, Zhou J, Yu W, Tong D, Chen
S(2009) Microfluidics: a new cosset for neurobiology. Lab Chip
9(5):644652
50. Myers FB, Lee LP (2008) Innovations in optical
microfluidictechnologies for point-of-care diagnostics. Lab Chip
8(12):20152031
51. Mauk MG, Ziober BL, Chen Z, Thompson JA, Bau HH
(2007)Lab-on-a-chip technologies for oral-based cancer screening
anddiagnostics: capabilities, issues, and prospects. Ann N Y
AcadSci 1098:467475
Microfabricated analytical systems for integrated cancer
cytomics 207
-
52. Weigl B, Domingo G, Labarre P, Gerlach J (2008) Towards
non-and minimally instrumented, microfluidics-based
diagnosticdevices. Lab Chip 8(12):19992014
53. Kiechle FL, Holland CA (2009) Point-of-care testing
andmolecular diagnostics: miniaturization required. Clin Lab
Med29(3):555560
54. Fernandes TG, Diogo MM, Clark DS, Dordick JS, Cabral
JM(2009) High-throughput cellular microarray platforms:
applica-tions in drug discovery, toxicology and stem cell
research.Trends Biotechnol 27(6):342349
55. Wheeler DB, Carpenter AE, Sabatini DM (2005) Cell
micro-arrays and RNA interference chip away at gene function.
NatGenet 37 Suppl:2530
56. Doroshow JH, Parchment RE (2008) Oncologic phase 0
trialsincorporating clinical pharmacodynamics: from concept
topatient. Clin Cancer Res 14(12):36583663
57. Rowinsky EK (2004) Curtailing the high rate of
late-stageattrition of investigational therapeutics against
unprecedentedtargets in patients with lung and other malignancies.
Clin CancerRes 10(12 Pt 2):42204226
58. Fesik SW (2005) Promoting apoptosis as a strategy for
cancerdrug discovery. Nat Rev Cancer 5(11):876885
59. Mita AC, Mita MM, Rowinsky EK (2004) Development
ofrationally designed, target-based agents for the treatment
ofadvanced colorectal cancer. Clin Colorectal Cancer
4(2):107123
60. Wlodkowic D, Skommer J, McGuinness D, Hillier C,
Darzyn-kiewicz Z (2009) ER-Golgi networka future target for
anti-cancer therapy. Leuk Res 33(11):14401447
61. Hyde BB, Twig G, Shirihai OS (2010) Organellar vs
cellularcontrol of mitochondrial dynamics. Semin Cell Dev
Biol.doi:10.1016/j.semcdb.2010.01.003
62. Tokimitsu Y, Kishi H, Kondo S, Honda R, Tajiri K, Motoki
K,Ozawa T, Kadowaki S, Obata T, Fujiki S, Tateno C, Takaishi
H,Chayama K, Yoshizato K, Tamiya E, Sugiyama T, Muraguchi A(2007)
Single lymphocyte analysis with a microwell array chip.Cytometry A
71(12):10031010
63. Tajiri K, Kishi H, Tokimitsu Y, Kondo S, Ozawa T, Kinoshita
K,Jin A, Kadowaki S, Sugiyama T, Muraguchi A (2007) Cell-microarray
analysis of antigen-specific B-cells: single cellanalysis of
antigen receptor expression and specificity. CytometryA
71(11):961967
64. Yamamura S, Kishi H, Tokimitsu Y, Kondo S, Honda R, RaoSR,
Omori M, Tamiya E, Muraguchi A (2005) Single-cellmicroarray for
analyzing cellular response. Anal Chem 77(24):80508056
65. Rettig JR, Folch A (2005) Large-scale single-cell trapping
andimaging using microwell arrays. Anal Chem 77(17):56285634
66. Lindstrm S, Eriksson M, Vazin T, Sandberg J, Lundeberg
J,Frisn J, Andersson-Svahn H (2009) High-density microwell chipfor
culture and analysis of stem cells. PLoS ONE 4(9):e6997
67. Lindstrm S, Mori K, Ohashi T, Andersson-Svahn H (2009)
Amicrowell array device with integrated microfluidic componentsfor
enhanced single-cell analysis. Electrophoresis 30(24):41664171
68. Thomas RS, Morgan H, Green NG (2009) Negative DEP trapsfor
single cell immobilisation. Lab Chip 9(11):15341540
69. Bocchi M, Lombardini M, Faenza A, Rambelli L, Giulianelli
L,Pecorari N, Guerrieri R (2009) Dielectrophoretic trapping
inmicrowells for manipulation of single cells and small
aggregatesof particles. Biosens Bioelectron 24(5):11771183
70. Voldman J, Gray ML, Toner M, Schmidt MA (2002)
AMicrofabrication-based dynamic array cytometer. Anal
Chem74:39843990
71. Sia SK, Whitesides GM (2003) Microfluidic devices
fabricatedin poly(dimethylsiloxane) for biological studies.
Electrophoresis24(21):35633576
72. Takayama S, Ostuni E, LeDuc P, Naruse K, Ingber
DE,Whitesides GM (2001) Subcellular positioning of small
mole-cules. Nature 411(6841):1016
73. Takayama S, Ostuni E, LeDuc P, Naruse K, Ingber
DE,Whitesides GM (2003) Selective chemical treatment of
cellularmicrodomains using multiple laminar streams. Chem Biol
10(2):123130
74. Klauke N, Smith G, Cooper JM (2009) Local
regionalstimulation of single isolated ventricular myocytes using
micro-fluidics. Anal Chem. doi:10.1021/ac9008429
75. Hung PJ, Lee PJ, Sabounchi P, Aghdam N, Lin R, Lee LP
(2005)A novel high aspect ratio microfluidic design to provide a
stableand uniform microenvironment for cell growth in a
highthroughput mammalian cell culture array. Lab Chip 5:4448
76. King KR, Wang S, Irimia D, Jayaraman A, Toner M, YarmushML
(2007) A high-throughput microfluidic real-time geneexpression
living cell array. Lab Chip 7:7785
77. Di Carlo D, Wu LY, Lee LP (2006) Dynamic single cell
culturearray. Lab Chip 6(11):14451449
78. Wang Z, Kim MC, Marquez M, Thorsen T (2007)
High-densitymicrofluidic arrays for cell cytotoxicity analysis. Lab
Chip 7:740745
79. Miyake M, Yoshikawa T, Fujita S, Miyake J (2009)
Transfectionmicroarray and the applications. Mol Biosyst
5(5):444449
80. Chan SD, Luedke G, Valer M, Buhlmann C, Preckel T
(2003)Cytometric analysis of protein expression and apoptosis
inhuman primary cells with a novel microfluidic chip-basedsystem.
Cytometry A 55(2):119125
81. Huh D, Gu W, Kamotani Y, Grotberg JB, Takayama S
(2005)Microfluidics for flow cytometric analysis of cells and
particles.Physiol Meas 26(3):7398
82. Wolff A, Perch-Nielsen IR, Larsen UD, Friis P, Goranovic
G,Poulsen CR, Kutter JP, Telleman P (2003) Integrating
advancedfunctionality in a microfabricated high-throughput
fluorescent-activated cell sorter. Lab Chip 3(1):2227
83. Wang MM, Tu E, Raymond DE, Yang JM, Zhang H, Hagen N,Dees B,
Mercer EM, Forster AH, Kariv I, Marchand PJ, ButlerWF (2005)
Microfluidic sorting of mammalian cells by opticalforce switching.
Nat Biotechnol 23(1):8387
84. Fu AY, Chou HP, Spence C, Arnold FH, Quake SR (2002)
Anintegrated microfabricated cell sorter. Anal Chem
74(11):24512457
85. Sugino H, Ozaki K, Shirasaki Y, Arakawa T, Shoji S, Funatsu
T(2009) On-chip microfluidic sorting with fluorescence
spectrumdetection and multiway separation. Lab Chip
9(9):12541260
86. Adams JD, Kim U, Soh HT (2008) Multitarget magnetic
activatedcell sorter. Proc Natl Acad Sci U S A
105(47):1816518170
87. Wlodkowic D, Skommer J, Faley S, Darzynkiewicz Z, CooperJM
(2009) Dynamic analysis of apoptosis using cyanine SYTOprobes: from
classical to microfluidic cytometry. Exp Cell
Res315(10):17061714
88. Takeda K, Jimma F (2009) Maintenance free biosafety
flowcy-tometer using disposable microfluidic chip
(FISHMAN-R).Cytometry B 76B:405406
89. Takao M, Jimma F, Takeda K (2009) Expanded applications
ofnew-designed microfluidic flow cytometer (FISHMAN-R).Cytometry B
76B;405
90. Chung TD, Kim HC (2007) Recent advances in
miniaturizedmicrofluidic flow cytometry for clinical use.
Electrophoresis 28(24):45114520
91. Watkins N, Venkatesan BM, Toner M, Rodriguez W, Bashir
R(2009) A robust electrical microcytometer with
3-dimensionalhydrofocusing. Lab Chip 9:31773184
92. Mao X, Lin SCS, Dong C, Huang TJ (2009) Single-layer
planaron-chip flow cytometer using microfluidic drifting based
three-dimensional (3D) hydrodynamic focusing. Lab Chip
9:15831589
208 D. Wlodkowic, J.M. Cooper
-
93. Kummrow A, Theisen J, Frankowski M, Tuchscheerer A,Yildirim
H, Brattke K, Schmidt M, Neukammer J (2009)Microfluidic structures
for flow cytometric analysis of hydrody-namically focussed blood
cells fabricated by ultraprecisionmicromachining. Lab Chip
9:972981
94. Rodriguez-Trujillo R, Castillo-Fernandez O, Garrido M,
Arun-dell M, Valencia A, Gomila G (2008) High-speed
particledetection in a micro-Coulter counter with
two-dimensionaladjustable aperture. Biosens Bioelectron
24:290296
95. Scott R, Sethu P, Harnett CK (2008)
Three-dimensionalhydrodynamic focusing in a microfluidic Coulter
counter. RevSci Instrum 79:046104
96. Zheng S, Liu M, Tai YC (2008) Micro coulter counters
withplatinum black electroplated electrodes for human blood
cellsensing. Biomed Microdevices 10:221231
97. Holmes D, Pettigrew D, Reccius CH, Gwyer JD, van Berkel
C,Holloway J, Davies DE, Morgan H (2009) Leukocyte analysisand
differentiation using high speed microfluidic single cellimpedance
cytometry. Lab Chip 9:28812889
98. Piacentini N, Demarchi D, Civera P, Knaflitz M (2008)
Bloodcell counting by means of impedance measurements in
amicrosystem device. Conf Proc IEEE Eng Med Biol Soc 4824-7
99. Wang X, Becker FF, Gascoyne PRC (2002) Membrane
dielectricchanges indicate induced apoptosis in HL-60 cells
moresensitively than surface phosphatidylserine expression or
DNAfragmentation. Biochim Biophys Acta 1564:412420
100. Cheung K, Gawad S, Renaud P (2005) Impedance
spectroscopyflow cytometry: on-chip label-free cell
differentiation. CytometryA 65:124132
101. Sohn LL, Saleh OA, Facer GR, Beavis AJ, Allan RS,
NottermanDA (2000) Capacitance cytometry: measuring biological
cellsone by one. Proc Natl Acad Sci U S A 97:1068710690
102. Atienza JM, Zhu J, Wang X, Xu X, Abassi Y (2005)
Dynamicmonitoring of cell adhesion and spreading on
microelectronicsensor arrays. J Biomol Screen 10:795805
103. Zhang H, Tu E, Hagen ND, Schnabel CA, Paliotti MJ, Hoo
WS,Nguyen PM, Kohrumel JR, Butler WF, Chachisvillis M, MarchandPJ
(2004) Time-of-flight optophoresis analysis of live whole cellsin
microfluidic channels. Biomed Microdevices 6:1121
104. Forster AH, Wang MM, Butler WF, Chachisvilis M, Chung
TD,Esener SC, Hall JM, Kibar O, Lykstad K, Marchand PJ, MercerEM,
Pestana LM, Sur S, Tu E, Yang R, Zhang H, Kariv I (2004)Use of
moving optical gradient fields for analysis of apoptoticcellular
responses in a chronic myeloid leukemia cell model.Anal Biochem
327:1422
105. Nerenberg M, Kariv I, McNeeley P, Marchand P, Sur S, Diver
J,Riccitelli S, Nieva J, Saven A (2006) Use of optophoresis as anin
vitro predictor of cell response to chemotherapy for
chroniclymphocytic leukemia. Leuk Lymphoma 47:21942202
106. Tamaki E, Hibara A, Tokeshi M, Kitamori T
(2003)Microchannel-assisted thermal-lens spectrometry for
microchipanalysis. J Chromatogr A 987:197204
107. Tamaki E, Hibara A, Tokeshi M, Kitamori T (2005)
Tunablethermal lens spectrometry utilizing microchannel-assisted
ther-mal lens spectrometry. Lab Chip 5:129131
108. Balslev S, Jorgensen AM, Bilenberg B, Mogensen KB,
Snake-nborg D, Geschke O, Kutter JP, Kristensen A (2006)
Lab-on-a-chip with integrated optical transducers. Lab Chip
6:213217
109. Chen P, Feng X, Du W, Liu BF (2008) Microfluidic chips for
cellsorting. Front Biosci 13:24642483
110. Kim HS, Son OT, Kim KH, Kim SH, Maeng S, Jung HI
(2007)Separation of apoptotic cells using a microfluidic
device.Biotechnol Lett 29(11):16591663
111. Xia N, Hunt TP, Mayers BT, Alsberg E, Whitesides
GM,Westervelt RM, Ingber DE (2006) Combined
microfluidic-micromagnetic separation of living cells in continuous
flow.Biomed Microdevices 8(4):299308
112. Qu BY, Wu ZY, Fang F, Bai ZM, Yang DZ, Xu SK (2008) Aglass
microfluidic chip for continuous blood cell sorting by amagnetic
gradient without labeling. Anal Bioanal Chem 392(78):13171324
113. Pamme N (2007) Continuous flow separations in
microfluidicdevices. Lab Chip 7(12):16441659
114. Pamme N (2006) Magnetism and microfluidics. Lab Chip
6(1):2438
115. Pamme N, Wilhelm C (2006) Continuous sorting of magnetic
cellsvia on-chip free-flow magnetophoresis. Lab Chip
6(8):974980
116. Lau AY, Lee LP, Chan JW (2008) An integrated
optofluidicplatform for Raman-activated cell sorting. Lab Chip
8:11161120
117. Yung CW, Fiering J, Muellerb AJ, Ingber DE (2009)
Micro-magneticmicrofluidic blood cleansing device. Lab
Chip9:11711177
118. Nagrath S, Sequist LV, Maheswaran S, Bell DW, Irimia D,
UlkusL, Smith MR, Kwak EL, Digumarthy S, Muzikansky A, Ryan P,Balis
UJ, Tompkins RG, Haber DA, Toner M (2007) Isolation ofrare
circulating tumour cells in cancer patients by microchiptechnology.
Nature 450(7173):12351239
119. Nevill JT, Cooper R, Dueck M, Breslauer DN, Lee LP
(2007)Integrated microfluidic cell culture and lysis on a chip. Lab
Chip7:16891695
120. Gomez-Sjoberg R, Leyrat AA, Pirone DM, Chen CS, Quake
SR(2007) Versatile, fully automated, microfluidic cell
culturesystem. Anal Chem 79:85578563
121. Chaw KC, Manimaran M, Tayad EH, Swaminathan S
(2007)Multi-step microfluidic device for studying cancer
metastasis.Lab Chip 7:10411047
122. Deutsch M, Deutsch A, Shirihai O, Hurevich I, Afrimzon
E,Shafrana Y, Zurgila N (2006) A novel miniature cell retainer
forcorrelative high-content analysis of individual untethered
non-adherent cells. Lab Chip 6:9951000
123. Revzin A, Sekine K, Sin A, Tompkins RG, Toner M
(2005)Development of a microfabricated cytometry platform
forcharacterization and sorting of individual leukocytes. Lab
Chip5:3037
Microfabricated analytical systems for integrated cancer
cytomics 209
Microfabricated analytical systems for integrated cancer
cytomicsAbstractIntroductionMicrofabricated systems for single
cytomicsLiving cell microarraysMicroflow chip-based
cytometryMicrofabricated cell sortersFuture outlookReferences
/ColorImageDict > /JPEG2000ColorACSImageDict >
/JPEG2000ColorImageDict > /AntiAliasGrayImages false
/CropGrayImages true /GrayImageMinResolution 150
/GrayImageMinResolutionPolicy /Warning /DownsampleGrayImages true
/GrayImageDownsampleType /Bicubic /GrayImageResolution 150
/GrayImageDepth -1 /GrayImageMinDownsampleDepth 2
/GrayImageDownsampleThreshold 1.50000 /EncodeGrayImages true
/GrayImageFilter /DCTEncode /AutoFilterGrayImages true
/GrayImageAutoFilterStrategy /JPEG /GrayACSImageDict >
/GrayImageDict > /JPEG2000GrayACSImageDict >
/JPEG2000GrayImageDict > /AntiAliasMonoImages false
/CropMonoImages true /MonoImageMinResolution 600
/MonoImageMinResolutionPolicy /Warning /DownsampleMonoImages true
/MonoImageDownsampleType /Bicubic /MonoImageResolution 600
/MonoImageDepth -1 /MonoImageDownsampleThreshold 1.50000
/EncodeMonoImages true /MonoImageFilter /CCITTFaxEncode
/MonoImageDict > /AllowPSXObjects false /CheckCompliance [ /None
] /PDFX1aCheck false /PDFX3Check false /PDFXCompliantPDFOnly false
/PDFXNoTrimBoxError true /PDFXTrimBoxToMediaBoxOffset [ 0.00000
0.00000 0.00000 0.00000 ] /PDFXSetBleedBoxToMediaBox true
/PDFXBleedBoxToTrimBoxOffset [ 0.00000 0.00000 0.00000 0.00000 ]
/PDFXOutputIntentProfile (None) /PDFXOutputConditionIdentifier ()
/PDFXOutputCondition () /PDFXRegistryName () /PDFXTrapped
/False
/Description > /Namespace [ (Adobe) (Common) (1.0) ]
/OtherNamespaces [ > /FormElements false /GenerateStructure
false /IncludeBookmarks false /IncludeHyperlinks false
/IncludeInteractive false /IncludeLayers false /IncludeProfiles
true /MultimediaHandling /UseObjectSettings /Namespace [ (Adobe)
(CreativeSuite) (2.0) ] /PDFXOutputIntentProfileSelector /NA
/PreserveEditing false /UntaggedCMYKHandling /UseDocumentProfile
/UntaggedRGBHandling /UseDocumentProfile /UseDocumentBleed false
>> ]>> setdistillerparams> setpagedevice