PAPER
`
`www.rsc.org/loc | Lab on a Chip
`
`Simultaneous measurement of reactions in microdroplets filled
`by concentration gradients†
`
`Nicolae Damean,a Luis F. Olguin,b Florian Hollfelder,b Chris Abella and Wilhelm T. S. Huck*a
`
`Received 27th November 2008, Accepted 23rd February 2009
`First published as an Advance Article on the web 19th March 2009
`DOI: 10.1039/b821021g
`
`This work describes a technology for performing and monitoring simultaneously several reactions
`confined in strings of microdroplets having identical volumes but different composition, and travelling
`with the same speed in parallel channels of a microfluidic device. This technology, called parallel
`microdroplets technology (PmD), uses an inverted optical microscope and a charge-coupled device
`(CCD) camera to collect images and analyze them so as to report on the reactions occurring in these
`microdroplets. A concentration gradient of one reactant is created in the microfluidic device. In each
`channel, a different concentration of this reactant is mixed with a fixed amount of a second reactant.
`Using planar flow-focusing methodology, these mixtures are confined in microdroplets of pL size which
`travel in oil as continuous medium, avoiding laminar dispersion. By analyzing the images of parallel
`strings of microdroplets, the time courses of several reactions with different reagent compositions are
`investigated simultaneously. In order to design the microfluidic device that consists in a complex
`network of channels having well-defined geometries and restricted positions, the theoretical concept of
`equivalent channels (i.e. channels having identical hydraulic resistance) is exploited and developed. As
`a demonstration of the PmD technology, an enzyme activity assay was carried out and the steady-state
`kinetic constants were determined.
`
`1.
`
`Introduction
`
`This paper describes a microsystem-based technology that
`compartmentalizes and measures simultaneously different reac-
`tions in pL volumes. This technology, which we call parallel
`microdroplets technology (PmD), analyses optically a set of
`reactions confined in strings of water-in-oil microdroplets that
`have identical volumes (5 to 60 pL), travel with identical speeds
`(0.25 to 6 mm s1) and contain different reactant concentrations.
`The technology provides a flexible method for analyzing
`a number of reactions simultaneously and continuously.
`Microdroplets technology is emerging as a robust technology
`for performing individual experiments in nL–pL volumes.1–4 This
`technology offers consistent compartmentalization (identical
`microdroplets under
`identical
`conditions), avoids
`cross-
`contamination of
`the samples to be analyzed by minimal
`manipulation of reactants and uses small quantities of reactants
`which are transported without dispersion. It has applications in
`chemistry and biology such as enzyme kinetics,5–9 cell based-
`assays,10,11 in vitro protein expression12 and mass production of
`monodisperse particles.13
`Despite the importance of performing and analyzing experi-
`ments simultaneously, no such technology is available for
`
`aDepartment of Chemistry, University of Cambridge, Lensfield Road,
`Cambridge, CB2 1EW, UK. E-mail: wtsh2@cam.ac.uk
`bDepartment of Biochemistry, Tennis Court Rd, Cambridge, CB2 1QW,
`UK
`† Electronic supplementary information (ESI) available: 1. Measurement
`of speed of microdroplets in the parallel channels (Table S1 and Movie
`S1); 2. construction of the standard curve (fluorescence vs. product
`concentration) for microdroplets in the parallel channels (Fig. S1). See
`DOI: 10.1039/b821021g
`
`microdroplets. Although the generation of parallel strings of
`microdroplets was demonstrated,14 this work is not suitable for
`carrying out different reactions and monitoring them simulta-
`neously. We developed the PmD based on microdroplets tech-
`nology, to perform simultaneous reactions. For this purpose, we
`generated a set of strings of microdroplets in the same device, so
`that each string of microdroplets carries out a reaction using
`a different concentration of reactants, and the device is easily
`interfaced to the outside world (sample loading and detection).
`We associated a certain time moment to each position in all
`channels, as all microdroplets travel steadily with the same speed.
`To monitor the reactions in microdroplets, we used an optical
`microscope coupled to a charged-coupled device (CCD) camera.
`In general, the PmD is useful for reactions of the following
`type:
`
`R1(c) + R2
`
`/ P(c,t)
`
`(1)
`
`where each string of microdroplets contains a certain concen-
`tration c of reactant R1, (c varies from 0 in the first string of
`microdroplets to its maximum value in the last string), and the
`same concentration of reactant R2. The concentration of the
`product, P, is time dependent t, but also dependent on c. Each
`string of microdroplets represents a certain time-dependent
`reaction described by Eq. (1), where c is constant. To generate
`a range of concentrations of reactant R1, we create gradients in
`microfluidics.
`In this paper, we establish the PmD technology and apply it to
`enzyme activity assays. This technology has further applications
`for chemical/biological assays in strings of pL-volumes, such as
`titration reactions, protein denaturation, protein-ligand inter-
`action, synthesis of microparticles, and in any other reactions,
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Lab Chip, 2009, 9, 1707–1713 | 1707
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
` / Journal Homepage
`
` / Table of Contents for this issue
`
`

`
`where the effect of concentration gradients of reactants is inter-
`rogated.
`
`2. Materials and methods
`
`We fabricated all microfluidic devices in poly(dimethylsiloxane)
`Sylgard 184 (PDMS) (Dow Corning), using soft lithography.15–17
`Briefly, we designed the microfluidic device by using Autocad
`2007 (AutoDesk) and a dark-field mask was printed (Circuit
`Graphics). Employing this mask in 1:1 contact photolithography
`with SU8-2025 negative photoresist (Microchem), we generated
`the negative master consisting of patterned photoresist on
`a silicon wafer (3-inches diam., Compart Technology). To drive
`fluids, we used electromechanical pumps (Harvard Apparatus
`PHD2000 Infusion) connected to the microfluidic device through
`polyethylene tubing PE20 (Becton Dickinson). For sealing the
`tubing to PDMS slabs, we used Araldite adhesive (Farnell). For
`all experiments, we generated aqueous microdroplets in mineral
`oil containing 2 wt % of non-ionic surfactant Span-80
`(Sigma-Aldrich). To study the reproducibility of gradient and
`microdroplets formation, we used food dyes (Tolbest) and red
`fluorescent protein (preparation described in18). For measuring
`the enzyme activity of E. coli alkaline phosphatase we used three
`stock solutions as follows: buffer (100 mM MOPS pH 7.9 with
`100 mM NaCl, 1.0 mM MgCl2 and 10 mM ZnSO2), 15 mM of
`fluorescein diphosphate (Invitrogen) in buffer, and 1.5 mM of
`E. coli alkaline phosphatase in buffer. Solutions of fluorescein
`diphosphate contained 1.125 mM of red fluorescent protein. The
`product of the enzyme reaction is fluorescein (Sigma-Aldrich).
`An inverted optical microscope (IX71, Olympus) served both
`as light source and detector. To collect color images, we used
`a CCD camera (Phantom Miro 3, Vision Research) connected to
`the microscope that was operated in the transmission mode. For
`the fluorescence measurements, we operated the same micro-
`scope in epifluorescence mode by using a mercury lamp (U-
`LH100HG, Olympus) as wide-field illumination. The sample was
`illuminated and the image collected by using the same objective
`(UPlanFL N 10x/0.30, Olympus). We separated the fluorescence
`emission from the illumination by using fluorescence mirror sets
`(U-MF2, U-MWIG3, Olympus). To acquire the emission light,
`we used an EM-CCD camera (iXon+, Andor Technologies) and
`analyzed these images with its accompanying software. To
`measure the emission light, we defined regions of interest (40-
`pixels length  10-pixels width), and recorded the emission light
`of these regions. We calculated the fluorescence intensity as
`a mean of 20 individual readings taken at 1-min time interval,
`each reading being an average of 5 measurements taken at 5-s
`time interval, after background subtraction.
`
`3. Results and discussion
`
`3.1. The concept of equivalent channels and design of the
`microfluidic device
`
`Our objective is to develop a technology that can generate,
`manipulate and analyze (bio)chemical reactions in nL–pL
`volumes, simultaneously. This technology essentially represents
`a miniaturised miniaturization of
`the multi-well plate (the
`established technology for simultaneous experiments on volumes
`in the range of mL-mL). One fundamental property of the
`
`microplate is that its wells have the same volume. Refining our
`objective for the PmD, we need to assure that our microdroplets
`have identical dimensions and flow with the same speed.
`We fabricated a microfluidic device on two layers (Fig. 1). The
`lower layer of the microfluidic device is detailed in Fig. 2. This
`device consists of the following sections:
`S1 (Generation of gradients of R1) – We produce solutions of
`R1 by using a technology to generate concentration gradients in
`laminar flow, based on diffusive mixing of substances.19 This
`section has two inlet ports, R1 and water. If the fluids loading
`these ports have identical flow rates, the flow rates are identical in
`all four channels at the end of S1, and we obtain the following
`concentrations of R1 in each channel: 0, 33, 67 and 100%, (i.e. the
`concentration varies linearly). All serpentine-shaped channels
`are identical in height, width and length.
`S2 (Generation of microdroplets and mixing) – We mix each
`solution of R1 generated in S1, with R2, both liquids in laminar
`flow, with a passive micromixer based on the Coanda effect that
`splits the fluid streams and recombines them.20 Then, we produce
`
`Fig. 1 3-D view of the microfluidic device demonstrating the parallel
`microdroplets (PmD) technology. Both layers of the device are presented.
`The inlet and outlet ports are marked distinctively. They were connected
`to the outside world through tubings. (A) Scheme. (B) Photo of the device
`placed on the microscope stage during experimental work.
`
`1708 | Lab Chip, 2009, 9, 1707–1713
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
`

`
`strings of microdroplets by using the planar flow-focusing
`methodology.21 All the microdroplets contain the same volu-
`metric percentage of R2, but different amounts of R1. A scale
`inset in Fig. 2 shows a quarter of S2, all quarters being identical.
`S3 (Lab and screening) – The reactions start and progress in all
`four channels, and the reactants are continuously mixed by
`chaotic advection induced in the microdroplets by serpentines.9,22
`To follow the reactions, images are taken (in transmission mode
`or epifluorescence mode) and analyzed. This section consists of
`identical channels, arranged in parallel to minimize the surface
`which is monitored by the camera as the dimension of the camera
`chips is a limiting factor for the applications. To facilitate the
`experimental work, these parallel channels are marked equidis-
`tantly in 1-mm steps. We inspect the images along these marks to
`collect the experimental data.
`S4 (Control and interface) – This section guaranties identical
`microdroplet size and speed in each channel, and also acts as the
`interface with the outside world for R2, oil, and the strings of
`microdroplets. This section is fabricated on two layers (Fig. 1).
`The upper layer contains two separate chambers. One chamber
`(the outlet chamber) collects all the four outlet ports from the
`lower layer in a single outlet port that connects by tubing to the
`outside world. The other chamber (the inlet chamber) distributes
`R2 (arriving to its inlet port via tubing) to all four inlet ports for
`R2 in the lower layer. These two chambers have large volumes, so
`they introduce negligible hydraulic resistances in our microfluidic
`device in comparison with the resistances introduced by the
`narrow and long channels in the lower layer. In this way, we can
`achieve identical pressure differences across all channels.
`Similar to the section S3, the channels for the strings of
`microdroplets in the section S4 should be identical in length,
`width, and height (Fig. 2). However, unlike section S3, these
`channels are kept widely apart to accommodate large outlet holes
`
`(1-mm-diameter, marked as outlet ports, and organized on
`a circle arc) between 50-mm-wide channels. A hole for the outlet
`port (punched in the outlet chamber of the upper layer) corre-
`sponds to the centre of this circle.
`The positioning of the channels (for the strings of micro-
`droplets) and of the holes between them prevented us from
`designing identical channels for reactant R2 and for oil in the
`section S4. Consequently, we have to use non-identical channels
`(i.e. channels having different geometries) but identical hydraulic
`resistance. For this purpose, we exploited and developed the
`theoretical concept of equivalent channels. This concept is
`applicable for rectangular and any other non-circular channels.
`Two channels are equivalent if they have identical hydraulic
`resistance (the fluids flow with identical flow rates when the
`pressure differences along these channels are identical). We detail
`next an analytical expression of this definition.
`For laminar flow, the hydraulic resistance of a channel is
`defined by:
`
`Rh ¼ dp
`Q
`
`(2)
`
`where dp is the pressure difference along the channel, Q is the
`flow rate defined by:
`
`Q ¼ Vav , p,D 2
`
`h
`
`4
`
`(3)
`
`Vav is the cross-sectional averaged velocity, and Dh is the
`hydraulic diameter of the channel.
`Using Eqs. (2) and (3), the hydraulic resistance is calculated as:
`
`Rh ¼ 8,Po
`
`p
`
`, m,dx
`D 4
`h
`
`(4)
`
`Fig. 2 Lower layer of the microfluidic device. The mask used for photolithography is displayed. Channels are coloured in accordance with their
`utilization: white for water, green for R1, red for R2, yellow for oil, and orange for the strings of microdroplets. The device is composed of four sections.
`In the section S1, a concentration gradient of R1 was produced in four parallel channels. Solutions of R1 were mixed with reactant R2 in the section S2,
`and strings of microdroplets having identical dimensions and flowing with the same speed were generated (see scale inset). These microdroplets were
`investigated using a CCD camera coupled to an optical microscope. The images were collected along the marks displayed in the section S3. The strings of
`microdroplets left the device in the section S4 through a single outlet port in the upper layer of the device (see also Fig. 1). Reactant R2 was distributed
`from a single inlet port of the upper layer into the lower layer of the device in the section S4 (see also Fig. 1). The oil was also distributed from a single
`inlet port. In the section S4, channels having different widths but identical heights were used for reactant R2 and for oil (see scale inset), in accordance
`with the concept of equivalent channels developed in this work.
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Lab Chip, 2009, 9, 1707–1713 | 1709
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
`

`
`where the dimensionless Poiseuille number Po is expressed in
`terms of experimental quantities as:
`
`Po ¼ 1
`m
`
`, dp
`dx
`
`, D 2
`2,Vav
`
`h
`
`m is the fluid viscosity and dp/dx is the pressure gradient along the
`channel.
`It has been proven that the Poiseuille number is solely
`dependent on the shape of the cross-section (e.g., rectangular,
`triangular, hexagonal).23 For rectangular channels (which are of
`interest for our design), Po depends only on the aspect ratio g of
`the cross-section, Po ¼ Po(g), where g ¼ h/w; h, w ¼ height and
`width of the channel, respectively. In this case, Po is expressed as
`a five-degree polynomial function of the argument g.24,25 The
`hydraulic diameter for rectangular channels is:
`
`Dh ¼ 2,h,w
`h þ w
`
`¼ 2,h
`g þ 1
`
`(5)
`
`Using Eq. (4), we exploited the definition of two equivalent
`rectangular channels as:
`Poðg1Þ,L1
`D 4
`h1
`
`¼ Poðg2Þ,L2
`D 4
`h2
`
`(6)
`
`and explicitly from Eqs. (5) and (6) as:
`Po(g1),L1,(g1 + 1)4 ¼ Po(g2),L2,(g2 + 1)4
`
`(7)
`
`Fig. 3 Two equivalent rectangular channels. The flow rates Q in these
`channels are identical if the pressure differences dp along these channels is
`identical, namely the channels have identical hydraulic resistance. This
`equivalence was used for designing the channels for reactant R2 and for
`oil, both in the lower layer of S4. These channels have the same heights h
`but different widths w and lengths L.
`
`where L is the length of the channel, and indexes are used to
`identify two equivalent rectangular channels having different
`lengths and widths, but identical height. For example, by using
`Eq. (7), two channels are equivalent if the width of the second is
`double the width of the first and the length of the second is 2.891
`times longer than the length of the first (Fig. 3).
`Applying the concept of equivalent channels for reactant R2
`and for oil, we obtained the design presented in Fig 2, where all
`ports in the lower layer of S4 are placed so that we can fabricate
`a compact device where all microdroplets have identical dimen-
`sions and flow with the same speed.
`From Eq. (7), we observe that the characteristic of two chan-
`nels as being equivalent does not depend on the fluid material
`properties, velocities, or temperatures, but it dependents only on
`the aspect ratio of the cross-sections and length of the channels.
`Also, for a given channel, there are infinite numbers of equivalent
`channels associated to it. For practical reasons (e.g., micro-
`fabrication restrictions, design flexibility), we chose to keep the
`same height for all channels, and also to work with only two
`dimensions for the width. A scale inset in Fig. 2 presents a small
`area of the lower layer of S4 to illustrate the utilization of
`channels having different widths.
`The modular structure of this microfluidic device and the
`technique of using equivalent channels in the S4 section, are
`powerful tools for future developments of this microfluidic
`device, e.g. to double the number of strings of microdroplets (and
`implicitly the number of reactions that are analyzed simulta-
`neously) from four to eight, and further, without difficulties.
`
`3.2. Fabrication and functionality
`
`We produced positive replicas in PDMS of the negative master
`containing 50-mm-deep channels by molding liquid pre-polymer
`and cross-linker in the recommended weight ratio of 10:1. After
`curing thermally (for 5 hours at 70 C), the layer of PDMS was
`cut and peeled away from the master. We punched the PDMS
`replica to create inlet and outlet ports for the fluids. The PDMS
`replica contained both the lower layer (S1, S2, S3, and part of S4),
`and the upper layer (the other part of S4). The upper layer was
`detached from the PDMS replica. Then, we sealed irreversibly
`the lower layer to a clean glass substrate, both being plasma
`oxidized (2 Torr, 60s, 100W) and brought together immediately
`after activation. Then, the upper layer was sealed irreversibly to
`the lower layer. We aligned these two layers by eye, with an
`
`Fig. 4 Demonstration of the PmD technology. Gradient formation, mixing of fluids, and microdroplets formation are shown. The flow rates were: 50 ml
`h1 for water and for green dye (R1), 75 ml h1 for red dye (R2), and 150 ml h1 for oil. The speed of microdroplets was 2.57 mm s1 in all channels.
`
`1710 | Lab Chip, 2009, 9, 1707–1713
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
`

`
`alignment error smaller than 0.5 mm which is suitable for our
`final target. We sealed with adhesive the tubing to the PDMS
`layers to prevent leaking.
`All the channels in this microfluidic device were 50-mm wide,
`excepting some parts of the channels that carry R2 and oil in the
`lower layer of S4 which are 100-mm wide, in accordance with the
`concept of equivalent channels.
`To check the functionality of the device, we used food dyes in
`Milli-Q water, namely green dye as R1 and red dye as R2 (Fig. 4).
`We generated gradients of green dye in Milli-Q water. Each
`string of microdroplets contained a certain concentration of
`green dye and a fixed concentration of red dye. We varied the
`flow rate of the aqueous streams from 0.5 ml h1 to 100 ml h1, and
`the flow rate of the oil stream from 25 mL h1 to 150 mL h1 to
`generate different sizes of microdroplets. For each set of condi-
`tions, we measured the length of the microdroplets in all four
`channels. We found that the deviation between these measure-
`ments was less than 4% for the microdroplets between 50 and 250
`mm. If the length of the microdroplets was less than 50 mm, this
`deviation was 9%. These results are in agreement with those
`reported in,14 where strings of microdroplets having identical
`composition were studied. For experiments that last a few hours,
`we preferred to work with microdroplets in the range from 100 to
`200 mm, because of their stability. In the Supplementary Infor-
`mation, we included results showing all microdroplets are trav-
`elling at the same speed under a large range of flow rates in the
`microfluidic device.
`
`3.3 Enzyme activity assays
`
`To demonstrate the use of PmD for biological applications, we
`measured the steady-state kinetics of the enzyme hydrolysis of
`fluorescein diphosphate by E. coli alkaline phosphatase. As R1
`we used a stock solution of fluorescein diphosphate in buffer. We
`loaded the water inlet with buffer to create a concentration
`gradient of R1 (0, 33, 67 and 100%) at the end of the section S1. In
`the section S2, we mixed individually each of these four streams
`of substrate with the stock solution of E. coli alkaline phospha-
`tase that plays the role of R2. We formed microdroplets con-
`taining these two reactants, and monitored, simultaneously, the
`progress of these four reactions by measuring the fluorescence
`emission along section S3.
`To make sure that our device worked as expected, we moni-
`tored formation of the gradients in the section S1, a task facili-
`tated by having red fluorescent protein in R1. Red fluorescent
`protein worked as a marker for our experiment because its
`emission wavelength (607 nm) is different from the emission
`wavelength (520 nm) of the fluorescein (product of our reaction).
`Therefore, we monitored the formation of the gradient without
`interfering with the reaction we studied (Fig. 5). We observed the
`fluids were fully mixed in the middle of the serpentines of the
`section S1.
`We measured the increase of fluorescence intensity as the
`microdroplets passed along section S3. With the flow rates used
`in this experiment, we were able to monitor the time course of the
`reaction for 10 seconds (Fig. 6A). Using the same device, we
`varied the flow rates of buffer, R1 and R2 (to keep their total flow
`rate constant, so all microdroplets had the same size and trav-
`elled with the same speed in all channels and for the duration of
`
`Fig. 5 Images collected in the measurement process of the enzyme
`activity assay. (A) Optical micrograph (transmission mode) of the first
`level of gradients formation, optical micrograph (fluorescence mode) of
`the concentration gradients, and the dependencies intensity vs. pixel
`number along two lines: green line for concentrations of 0 and 100%, and
`red line for concentrations of 0, 50, and 100%. (B) Similar with (A), but
`for the second level of gradients: green line for concentrations of 0, 50,
`and 100%, and red line for concentrations of 0, 33, 67, and 100%. The
`concentration levels (indicated by stars) varied linearly across the device.
`(C) Optical micrographs (fluorescence mode) of the microdroplets
`carrying the enzyme reaction. These images were taken along the marks
`in the section S3 (see also Fig. 2). Red frames indicate the regions of
`interests monitored by the camera.
`
`the entire experiment) and we collected two further sets of data
`(Figs. 6B and 6C). In this manner, we obtained three sets of data
`associated to different ratios of substrate concentration vs.
`enzyme concentration. To convert the fluorescent measurements
`to product concentrations we used a standard curve constructed
`by operating the device with buffer, product and enzyme (see
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Lab Chip, 2009, 9, 1707–1713 | 1711
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
`

`
`Fig. 6 Enzyme activity assays. (A)–(C) Fluorescent intensity as a function of time measured by PmD for the hydrolysis reaction of fluorescein
`diphosphate by E. coli alkaline phosphatase at pH 7.9. Each graph corresponds to an experiment performed at fixed flow rates of oil, buffer, fluorescein
`3.33 mM and : 0.0 mM; enzyme concen-
`10 mM,
`6.67mM,
`diphosphate and E. coli alkaline phosphatase. (A) Initial substrate concentrations:
`tration: 0.5 mM; flow rates: 10, 10, 10, 10 (for oil, buffer, fluorescein diphosphate and E. coli alkaline phosphatase respectively, expressed in ml$h1). (B)
`2.67 mM and : 0.0 mM; enzyme concentration: 0.7 mM; flow rates: 10, 8, 8, 14. (C) Initial
`8.0 mM,
`5.33 mM,
`Initial substrate concentrations:
`1.67 mM and : 0.0 mM; enzyme concentration: 1.0 mM; flow rates: 10, 5, 5, 20. In all graphs from (A)–
`5.0 mM,
`3.33 mM,
`substrate concentrations:
`(C), the error bars represent standard deviation of 100 measurements. (D) Michaelis-Menten graph of the hydrolisis of fluorescein diphosphate by E. coli
`alkaline phosphatase obtained from the normalized experimental data presented in Fig. 6A (blue stars); 6B (red stars) and 6C (green stars). The R2 of the
`fitting quality is 0.98. Inset presents comparison between the steady-state kinetic constants obtained by the PmD and a conventional 96-well plate.
`
`for each data set, we
`Supplementary Information). Then,
`calculated the initial rates and we normalized them by dividing to
`the enzyme concentration. We plotted the normalized initial
`rates vs. substrate concentration and fitted to the Michaelis-
`Menten equation vn ¼ kcat
`$[R1]$/(KM + [R1]), where vn is the
`normalized initial rate, [R1] is the initial substrate concentration
`for each experiment, kcat is the turnover rate constant, and KM is
`the Michaelis constant (apparent dissociation constant of the
`enzyme-substrate complex). We obtained values of kcat and KM
`(Fig. 6D). These values are in close agreement with data obtained
`when we performed a conventional assay in a 96-well plate using
`the same reactants as those used in the microfluidic device.
`The enzyme assay studied in this paper is an example of using
`the PmD to perform simultaneously a set of chemical/biochemical
`reactions.
`A major advantage is that we only need to load our device with
`four liquids to run the entire study, and no further dilutions are
`needed. By following the same approach, a device providing
`
`a gradient of eight or more different concentrations would be
`straightforward to make. Also, in the PmD technology, we can
`use concentrations that vary non-linearly if the loading fluids
`have different flow rates or/and we change the geometry of S1.19
`Concentration gradients can be beneficial for the analysis of
`biological processes such as axon pathfinding,26 high content
`screening of cells,27 chemotaxis studies,28,29 high-throughput cell-
`based assays,30 and monitoring of gene expression in live cells.31
`This technology minimizes the costs of the samples, contami-
`nation of reagents, and also the manual labour involved in
`analysing larger number of samples. By manipulation a few
`syringes during an experimental session, it is possible to modify
`continuously the concentration of reactants in microdroplets,
`broaden or focus the concentration ranges studied based on the
`results obtained so far in that session. The time range studied
`(and thus the sensitivity of the assay) can be adjusted by
`extending the length of the microfluidic channels along which
`images are recorded.
`
`1712 | Lab Chip, 2009, 9, 1707–1713
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Online
`
`

`
`4. Conclusions
`
`This work demonstrates a versatile and reliable technology for
`compartmentalization and simultaneous monitoring of different
`reactions in parallel strings of microdroplets generated in
`microsystems. The parallel microdroplets technology (PmD)
`integrates readily available equipment (an optical microscope
`and a CCD camera) with a microfluidic device. Minimal
`manipulation of the fluids is needed, because this device has only
`three inlet ports for the aqueous phase (two reactants and water)
`to generate simultaneously a set of reactions in microdroplets
`flowing in an oil phase. The design of the device used for this
`technology is possible thanks to the concept of equivalent
`channels which proves to be an important design principle for
`PmD technology
`complex microfluidic
`networks. The
`provides a practical capability to the rapidly growing field of
`microanalysis.
`
`Acknowledgements
`
`This work was supported by the EPSRC and the RCUK Basic
`Technology Programme.
`
`References
`
`1 T. Thorsen, R. W. Roberts, F. H. Arnold and S. R. Quake, Phys. Rev.
`Lett., 2001, 86, 4163–4166.
`2 S.-T. Teh, R. Lin, L.-H. Hung and A. P. Lee, Lab Chip, 2008, 8, 198–
`220.
`3 B. T. Kelly, J.-C. Baret, V. Taly and A. D. Griffiths, Chem. Commun.,
`2007, 1773–1788.
`4 H. Song, D. L. Chen and R. F. Ismagilov, Angew. Chem., Int. Ed.,
`2006, 45, 7336–7356.
`5 H. Song and R. F. Ismagilov, J. Am. Chem. Soc., 2003, 125, 14613–
`14619.
`6 L. S. Roach, H. Song and R. F. Ismagilov, Anal. Chem., 2005, 77,
`785–796.
`7 E. M. Miller and A. R. Wheeler, Anal. Chem., 2008, 80, 1614–
`1619.
`8 K. Ahn, J. Agresti, H. Chong, M. Marquez and D. A. Weitz, Appl.
`Phys. Lett., 2006, 88, 264105.
`
`9 A. Liau, R. Karnik, A. Majumdar and J. H. D. Cate, Anal. Chem.,
`2005, 77, 7618–7625.
`10 A. Huebner, L. F. Olguin, D. Bratton, G. Whyte, W. T. S. Huck,
`A. J. deMello, J. B. Edel, C. Abell and F. Hollfelder, Anal. Chem.,
`2008, 80, 3890–3896.
`11 J. Clausell-Tormos, D. Liber, J.-C. Baret, A. El-Harrak, O. J. Miller,
`L. Frenz, J. Bouwolff, K. J. Humphry, S. K€oster, H. Duan, C. Holtze,
`D. A. Weitz, A. D. Griffiths and C. A. Merten, Chemistry & Biology,
`2008, 15, 427–437.
`12 F. Courtois, L. F. Olguin, G. Whyte, D. Bratton, W. T. S. Huck,
`C. Abell and F. Hollfelder, ChemBioChem, 2008, 9, 439–446.
`13 T. Nisisako and T. Torii, Lab Chip, 2008, 8, 287–293.
`14 W. Li, E. W. K. Young, M. Seo, Z. Nie, P. Garstecki, C. A. Simmons
`and E. Kumacheva, Soft Matter, 2008, 4, 258–262.
`15 D. C. Duffy,
`J. C. McDonald, O.
`J. A. Schueller and
`G. M. Whitesides, Anal. Chem., 1998, 70, 4974–4984.
`16 V. Linder, H. K. Wu, H. K. Jiang and G. M. Whitesides, Anal. Chem.,
`2003, 75, 2522–2527.
`17 S. K. Sia and G. M. Whitesides, Electrophoresis, 2003, 24, 3563–3576.
`18 R. E. Campbell, O. Tour, A. E. Palmer, P. A. Steinbach, G. S. Baird,
`D. A. Zacharias and R. Y. Tsien, Proc. Natl. Acad. Sci. USA, 2002,
`99, 7877–7882.
`19 S. K. W. Dertinger, D. T. Chiu, N. L. Jeon and G. M. Whitesides,
`Anal. Chem., 2001, 73, 1240–1246.
`20 C.-C. Hong, J.-W. Choi and C. H. Ahn, Lab Chip, 2004, 4, 109–113.
`21 S. L. Anna, N. Bontoux and H. A. Stone, Appl. Phys. Lett., 2003, 82,
`364–366.
`22 H. Song, J. Tice and R. F. Ismagilov, Angew. Chem., Int. Ed., 2003,
`42, 768–772.
`23 R. K. Shah and A. L. London, Laminar flow forced convection in
`ducts, Academic Press, New York, 1978.
`24 N. Damean and P. P. L. Regtien, Sensors. Actuators A, 2001, 90, 96–
`101.
`25 W. M. Rohsenow, J. P. Harnett and Y. I. Cho, Handbook of Heat
`Transfer, 3rd Ed., McGraw-Hill, New York, 1998.
`26 S. K. W. Dertinger, X. Y. Jiang, Z. Y. Li, V. N. Murthy and
`G. M. Whitesides, Proc. Natl. Acad. Sci. USA, 2002, 99, 12542–12547.
`27 N. Ye, J. Qin, W. Shi, X. Liu and B. Lin, Lab Chip, 2007, 7, 1696–
`1704.
`28 G. M. Walker, J. Sai, A. Richmong, M. Stremler, C. Y. Chung and
`J. P. Wikswo, Lab Chip, 2005, 5, 611–618.
`29 D. Irimia, S.-Y. Liu, W. G. Tharp, A. Samadani, M. Toner and
`M. C. Poznansky, Lab Chip, 2006, 6, 191–198.
`30 P. J. Hung, P. J. Lee, P. Sabounchi, R. Lin and L. P. Lee, Biotechnol.
`Bioeng., 2005, 89, 1–8.
`31 D. M. Thompson, K. R. King, K. J. Wieder, M. Toner,
`M. L. Yarmush and A. Jayaraman, Anal. Chem., 2004, 76, 4098–4103.
`
`This journal is ª The Royal Society of Chemistry 2009
`
`Lab Chip, 2009, 9, 1707–1713 | 1713
`
`Published on 19 March 2009. Downloaded by Radboud Universiteit Nijmegen on 15/04/2016 15:48:58.
`
`View Article Onl