Microfluidic system

Abstract

The present invention concerns a microfluidic system comprising: a microchannel containing several elements of two non-miscible fluids, the microchannel comprising a droplet (30) containing magnetic particles (M), and a device for generating inside the microchannel magnetic field, said device comprising an activable magnetic element, the activable magnetic element comprising a tip (5,6), the microfluidic system being configured to transport the droplet by flow or by pressure difference.

Claims

1. A microfluidic system comprising: a microfluidic device comprising a microchannel, in the microchannel, a sequence of droplets of at least one first fluid in a surrounding immiscible second fluid, at least one droplet containing magnetic particles being transported by a flow of said second fluid through a portion of the microchannel; a device for generating inside the portion of the microchannel a magnetic field, said device comprising an activable magnetic element, activation of the activable magnetic element creating the magnetic field in the portion of the microchannel causing the magnetic particles to be captured in said portion, wherein the activable magnetic element comprises: a core that is reversibly magnetizable, and a tip made out of a soft magnetic material and having an acute tip angle in one direction only, the tip being magnetically coupled to the core, the tip having a cross-sectional area decreasing towards the microchannel, the decreasing cross-sectional area creating a convergence of magnetic field lines inside said tip, wherein the end of the tip proximal the microchannel has a longitudinal dimension greater than a width of the microchannel.

2. The system according to claim 1, wherein the core is surrounded by a conducting coil connected to a current generator.

3. The system according to claim 1, further comprising a plurality of activable magnetic elements.

4. The system according to claim 1, wherein the magnetic particles do not comprise ferromagnetic particles.

5. The system according to claim 1, wherein the magnetic particles are surface-functionalized.

6. The system according to claim 1, wherein the magnetic particles form an aggregate of magnetic particles.

7. The system according to claim 1, the microchannel containing a plurality of droplets containing magnetic particles.

8. The system according to claim 1, wherein the device for generating the magnetic field inside the microchannel creates inside the microchannel a magnetic field having field lines which are not collinear to the longitudinal axis of the microchannel so as to capture the magnetic particles.

9. The system according to claim 1, further comprising an acoustic wave generator for exposing the inside of the microchannel to ultrasonic waves.

10. The system according to claim 1, wherein the microfluidic system further comprises a syringe pump.

11. The system according to claim 1, wherein the device comprises additional activable magnetic elements which face the activable magnetic element across the microchannel.

12. The system according to claim 1, wherein the end of the tip of the activable magnetic element proximal the microchannel is a flat end.

13. The system according to claim 12, wherein the flat end of the tip occupies an area that is smaller than the rest of the tip.

14. The system according to claim 13, wherein the area of the flat end of the tip occupies less than 2% of the area of the rest of the tip.

15. The system according to claim 1, wherein the acute tip angle of the activable magnetic element is smaller than 45.

16. The system according to claim 1, wherein the end of the tip proximal the microchannel has a dimension measured along a longitudinal axis of the microchannel that is less than a length of one of the droplets in the sequence of droplets.

17. The system according to claim 16, wherein the ratio of the dimension of the end of the tip to the length of the droplet is less than 1:5.

18. The system according to claim 1, wherein the microchannel passes several times in the vicinity of the convergence of magnetic field lines.

19. The system according to claim 1, wherein the activable magnetic element is one of a pair of activable magnetic elements which face each other across the microchannel.

20. An assembly comprising: the microfluidic system according to claim 1, and a detector for measuring at least one characteristic of a droplet containing magnetic particles present in the microchannel.

21. A method of manipulating magnetic particles, the method using the microfluidic system according to claim 1 and comprising: capturing and aggregating the magnetic particles contained in a droplet in the vicinity of at least one activable magnetic element by submitting the magnetic particles to a magnetic field generated by the at least one activable magnetic element, wherein the at least one activable magnetic element creates conditions chosen from a magnetic field intensity being comprised between 10 mT and 1 T and a magnetic field gradient being comprised between 10 and 10000 T/m along the longitudinal axis of the microchannel, the magnetic particles being surface functionalized, each of the surface functionalized magnetic particles providing at least one binding site for a target, the target being present in the droplet containing the magnetic particles.

22. A method of extracting at least one magnetic particle from a first primary droplet of a first fluid flowing in a microchannel, said method using the microfluidic system according to claim 1 and comprising: capturing the at least one magnetic particle in the vicinity of at least one activable magnetic element by submitting the at least one magnetic particle to a magnetic field generated by the at least one activable magnetic element, deforming the first primary droplet, and splitting the first primary droplet into a first secondary droplet and a second secondary droplet, the first secondary droplet comprising the at least one magnetic particle and remaining captured by the magnetic field in the vicinity of the at least one activable magnetic element to extract the at least one particle from the first primary droplet.

23. A method of manipulating at least one magnetic particle comprising: extracting the at least one magnetic particle from a first primary droplet of a first fluid flowing in a microchannel according to the method defined in claim 22, and releasing the at least one magnetic particle by modifying an intensity and/or gradient of the magnetic field created by the at least one activable magnetic element.

Description

BRIEF DESCRIPTION OF THE DRAWINGS

(1) FIG. 1 is a partial diagrammatic view of a microfluidic device according to the invention,

(2) FIG. 1A is a diagrammatic view of a microfluidic device according to the invention,

(3) FIG. 2 diagrammatically represents the different steps of a bead-handling method according to the invention,

(4) FIGS. 3A and 3B show phase diagrams illustrating the conditions required for bead-cluster extraction,

(5) FIG. 4 show an example of method according to the invention,

(6) FIGS. 5 to 8 show the results obtained by a method according to the invention,

(7) FIG. 9 diagrammatically represents the different steps of another bead-handling method according to the invention,

(8) FIG. 10 shows bright field micrographs of agglutination of magnetic beads in droplets,

(9) FIG. 11 shows the influence of intensity of the magnetic field on the method depicted in FIG. 9, and

(10) FIGS. 12 to 14 and 15 to 15F diagrammatically show details of variants of the microfluidic device that can be used in the methods according to the invention,

(11) FIGS. 16A and 16B show examples of activable magnetic element, and

(12) FIG. 17 show an embodiment of a micro channel according to the invention.

(13) The extraction can be successfully performed with only one magnetic tip connected to an electromagnetic coil. Here is a schematic illustration describing the tip geometry (coil implantation in dashed line):

(14) The materials used in our systems were mu-Metal or AFK502 or AFK01 (Arcelor Metal). We used a home made cylindrical electromagnetic coil composed of around 1000 turns of insulated copper wire (0.8 mm diameter). The current intensity used ranges from 0 up to 4 A.

(15) A second tip, facing the first one can be added to create a pair of magnetic activable elements.

(16) This second tip helps in focalizing the magnetic field lines and thus increases the magnetic field gradient in the microfluidic channel. As a consequence droplet splitting can be performed at lower magnetic particle content or at lower field intensity (i.e. lower current).

(17) In another configuration, the second tip can be connected to an additional electromagnetic coil.

(18) This can be done to increase the field intensity and thus to facilitate even more droplet splitting and particle extraction.

(19) FIG. 1 shows a microfluidic system 1 comprising a micro channel 2 extending along a longitudinal axis X and a capture device 3 for generating a magnetic field to capture magnetic bodies and perform a method according to the invention, for example as illustrated on FIG. 2.

(20) The device 3 comprises at least one activable magnetic element 5 and in the example shown comprises two activable magnetic elements 5 and 6. The pair of elements 5 and 6 is also referred to as magnetic tweezers.

(21) The magnetic elements 5 and 6 face each other and extend along an axis Y that is transverse to the axis X, preferably perpendicular to the axis X. Magnetic elements 5 and 6 each comprise a corresponding tip 5a and 6a having a cross-section perpendicular to axis Y that decreases toward the other tip. The cross-section of the tips 5a and 6a decreases when moving towards the microchannel 2 as shown.

(22) In the example shown, each tip has an elongated cross-section and a respective edge 5b or 6b extending in a plane perpendicular to the axis X.

(23) The two edges 5b and 6b may extend parallel to each other, as shown, being spaced by a distance that for example does not exceed five times the thickness of the micro channel in the direction Y.

(24) The edges 5b and 6b may contact walls defining the micro channel, while being isolated by the walls from fluids circulating inside the micro channel.

(25) The width of the tips 5a and 5b may exceed the width of the micro channel, as shown.

(26) The tips 5a and 6a are preferably made out of a soft magnetic material such as metal so as to exhibit no or little remanent magnetism after excitation.

(27) The tips 5a and 6a may be made integrally with a corresponding core 5c or 6c.

(28) An excitation coil 12 is wounded around core 5c to generate an induction collinear with axis Y.

(29) As shown in FIG. 1A, the coil 12 is connected to a controller 15 that may comprise a computer running a software configured to excite the coil 12 based on various input data and for example based on optical droplet detection.

(30) A camera 16 may monitor the micro channel and provide the controller 15 with droplet detection information. This information enables automatic excitation of the coil 12 when a droplet is in the vicinity of the magnetic element 5 to perform for example the method illustrated in FIG. 2.

(31) The system of the invention also comprises a microfluidic device 20 for generating droplets in the micro channel 2.

(32) The device 20 comprises for example means forming droplets as disclosed by Chabert et al. in [2].

(33) The device 20 may also be able to generate droplets via a Taylor cone phenomenon as disclosed in WO 2004/091763. Both publications are hereby incorporated by reference.

(34) When exciting the coil 12, a magnetic field is generated by tip 5a and the magnetic field extends in the gap between tips 5a and 6a. The field is intense between the tips 5a and 6a and quite narrow which enables to capture magnetic beads within a droplet as shown in FIG. 2.

(35) The captured magnetic beads are superparamagnetic particles.

(36) The controller 15 is configured for periodic demagnetization of the activable magnetic elements 5 and 6 by powering the coil 12 with an alternative current of decreasing magnitude.

(37) The activable magnetic element 5 is the only one in the example shown in FIG. 1 that is provided with a coil. In a variant non shown, the opposite element 6 is also provided with an excitation coil.

(38) The coil 12 may be excited by a DC or AC current.

(39) Preferably, as shown, the coil 12 is spaced from the edge of the tip, for example by at least 1 cm or 2 cm, which may reduce transfer of heat from the coil toward the micro channel.

(40) The obtained experimental results are hereunder detailed.

EXAMPLES

Example 1: Particle Extraction

(41) Experiments were performed with two kinds of devices. Device 1 was composed of a single magnetic tip made of soft magnetic alloys (AFK502R, Imphy Alloys Arcelor Mittal). An electromagnetic coil (33.5 mm diameter) made of 1000 loops of copper wire is used to control magnetization of one tip. The magnetic tip is placed perpendicularly with a Teflon tubing (300 m ID and 600 m OD, Sigma-Aldrich).

(42) We used a home made cylindrical electromagnetic coil composed of around 1000 turns of insulated copper wire (0.8 mm diameter). The current intensity used ranges from 0 up to 4 A.

(43) In a second type of device (device 2), a second tip, facing the first one can be added to create a magnetic tweezers configuration (reflection symmetry).

(44) The different steps of the carried out method are shown in FIG. 2.

(45) Droplets trains were produced in the Teflon tubing using an automated pipeting robot system.

(46) Water droplets 30 were generated in fluorinated oil 40 (FC-40, 3M) with 3% of surfactant (1H,1H,2H,2H-perfluorodecan-1-ol, Fluorochem). Considering 100 nL droplets, we introduced 1 g of particles M (1 m, Dynal MyOne COOH Particles)

(47) The system is composed of an aspirating tip (Teflon tubing Sigma) carried by a XYZ displacement stage and connected to a NeMesys syringe pump system (Cetoni), both controlled with Labview (National Instrument). Samples are stored in microtiter plate. In each well, aqueous samples are covered with fluorinated oil. The system sequentially aspirates aqueous solution and fluorinated oil in each well in order to produce a custom train of 80 nL droplets 30 with a spacing of 150 nL. During the droplets formation, the tube containing the magnetic particles suspension is mixed at least every 5 minutes in order to avoid beads sedimentation.

(48) The typical velocity of the droplets trains was 1 mm/s. The oil flow rate was kept constant during all the experiment.

(49) When passing through the magnetic tweezers 5 and 6, the current intensity in the electromagnet was switched from 0 up to 1.5 A for device 2 and from 0 to 2 A for device 1. The resulting magnetic field induces an attracting force toward the active tip. A magnetic bead cluster 35 is generated. While the droplet 30 is passing in between the tweezers, the cluster 35 is kept immobile until it reaches the water/oil interface of the droplet. When reaching the leading edge of the droplet, the magnetic pulling force acting on the bead-cluster is transferred to the water/oil interface and deforms the droplet. Above a certain threshold, the magnetic force overcomes capillary force and a particle-containing droplet 31 is split out of the mother droplet. The mother droplet 32 is dragged away by the oil flow while the extracted droplet 31, is retained in the trap. Obviously, this condition is fullfilled if the size of the resulting droplet, which is mainly dictated by the volume of the particle cluster, is smaller than the inner capillary size. In our experiments, the volumes of the extracted droplet was 1 nL.

(50) The cluster-droplet 31 can be kept confined in the magnetic tweezers 5 and 6 until a new plug 33 comes into contact and merges spontaneously.

(51) These different steps are diagrammatically shown in FIG. 2.

(52) After merging, depending on the operation to be performed, the magnetic field can be:

(53) a) maintained active to keep particles trapped in the trap while the droplet is passing.

(54) b) released in order to re-suspend particles in a new droplet 34. In the latter case, the recirculating flows created by droplet motion give rise to internal hydrodynamic recirculation flows that enhances sample and particle mixing.

(55) This strategy offers all the basic functions required for a bioassay namely particle washing, extraction and incubation in sub-microliter sample volumes. Confined droplets manipulation is advantageous as they offer small volumes, easy spatial and temporal handling and give rise to internal hydrodynamic recirculation flows that enhances sample mixing. As compared to earlier work, the use of a fluorinated-oil in fluorocarbon capillaries avoids contamination associated with contact between the water plug and the surface. Droplet containing magnetic particles can be manipulated individually, with very short response time, using electromagnetically actuated tweezers. This new approach allows a better extraction efficiency and lower sample volumes than reported e.g. in [3,4], and the easy and flexible implementation of complex protocols, when combined with a robotized drop formation platform [2] allowing the formation of trains comprising any number of droplets arbitrarily selected from a microtiter plate.

(56) Several details concerning the method according to the invention are provided hereunder.

(57) Capture Characterization

(58) The magnetic force Fm used to extract a magnetic-particle cluster from the drop is related to the magnetic field gradient:

(59) $F_m=\left(\frac{Q}{}\right)\frac{B}{{}_0}B$

(60) Here Q is the mass of magnetic particles, the mass density, the bead magnetic susceptibility, 0 the permittivity of free space and B the magnetic flux density. The magnetic field gradient in the tubing may be influenced by the geometry of the tip and more particularly by its sharpness.

(61) The adding of a second tip on the other side of the tubing may enable to concentrate the field lines. Consequently, a higher local value of the gradient field may be achieved in the tubing even if the second tip is not magnetized by a coil.

(62) When magnet beads are attracted by tips, they produce a pressure on the droplet side resulting in a surface deformation. The bead extraction results from the balance between the magnetic and the capillary force. Fm has to be strong enough to overcome the capillary force created by the droplet surface deformation (FIG. 2) in order to break the surface. The capillary force Fc is related to the surface tension between water and oil estimated at 10 mN.Math.m1 (Dorfman et al., 2005) [13] and the size of the bead cluster. The force is given by (Shikida et al., 2006).

(63) $F_C=6^{\frac{1}{2}}{}^{\frac{2}{3}}{\left(\frac{Q}{}\right)}^{\frac{1}{2}}$
Conditions Required for Bead-Cluster Extraction

(64) The relative influence of current intensity in the magnetic coil and particle load on bead-cluster extraction was investigated for a droplet velocity of about 300 m/s which corresponds to a flow rate of 0.02 L/s. The results are shown in FIG. 3A.

(65) The relative influence of current intensity in the magnetic coil and droplet velocity on bead-cluster extraction was investigated for a particle loading of 0.8 g/droplet. The results are shown in FIG. 3B.

(66) We observed a small influence of the droplet speed on the ability to capture magnetic beads.

(67) When the drop is moving too fast, the time to create the beads cluster close by the tips is reduced.

(68) In some cases, the cluster formed does not contain the total amount of beads. This can have several consequences ranging from a decrease in the capture efficiency: some beads remain in the drop to the loss of extraction because the magnetic force necessary to extract beads is not able to overcome the capillary force. For high throughput application, the flow is directly related to the measurement capacity of the system. Then, the magnetic has to be tuned to be sure that all beads are extracted.

(69) Merging and Mixing

(70) Once the beads are extracted, the small droplet created is composed of magnetic particles and residual liquid. By measuring the droplet size and knowing the bead quantity in the cluster, the volume of the residual liquid is valued at 1 nL for initial 80 nL droplets containing 0.8 g of beads.

(71) As soon as the small cluster is in contact with another droplet, the merging always takes place.

(72) Indeed the two droplets coalesce when the oil film between surfaces is thin enough and beads are suspended in without forming aggregates. Then magnetic particles are dispersed in the liquid leading to an 80-fold dilution factor for the residual liquid. Consequently, washing steps required for immunoassay can be performed only by merging and capturing successively beads in a buffer droplet which takes less than 1 s.

(73) When particles are suspended from a cluster to a droplet in movement, they are compacted at the droplet end in a triangle-shaped area. However the bead mixing is efficient because plugs moving in tubing without wetting walls generate recirculating flow in oil and consequently in the drop too (Song et al., 2003) [14]. By this way, beads are dragged along those flow lines which lead to an active mixing in the entire droplet (FIG. 2).

(74) Beads Capture Efficiency

(75) We have discussed above the beads extraction driving forces and conditions. Here, we focus on the capture efficiency of our system. In order to quantify beads, fluorescent magnetic particles were prepared by saturating streptavidin coated beads with biotin atto 550. FIG. 8 shows the fluorescence profile of each plug after the capture and release processes.

(76) Because of the compaction of magnetic beads at the end drop and because fluorescence observation is done on the back of the tubing, the intensity level reaches a constant maximum value in the middle of this area. As the MPs quantity is more related to the size of this area than the maximum fluorescence intensity, the fluorescence is integrated over the drop length. In this way, it becomes possible to discriminate the quantity of beads contained in each droplet. But the relation between the integrated fluorescence and MPs amount is not directly proportional that is why ratios between initial and final drops fluorescence are not conserved. As the MPs compaction leads to the under-estimation of the real quantity of beads, the comparison of the fluorescence signal from initial drops and empty ones give us a capture efficiency higher than 99% which is actually an under the real value. The remaining magnetic particles seem to be trapped at the meniscus at the drop end. However this capture efficiency value is high enough to plan to use this system in a multi-step implementation such as sandwich immunoassay without losing biological materials.

Example 2: Immunoassay within Droplets

(77) As previously described (example 1) the basic operation units required for an immunoassay can be performed using the droplet platform: beads confinement, beads washing, beads release and mixing in a given droplet as well as continuous fluorescence monitoring.

(78) The immunoassay developed in this example is a sandwich immunoassay with capture antibody grafted on magnetic particles (from micro to nanoparticles). Secondary antibody (detection antibody) can be fluorescently labelled (FITC, Alexa . . . ) or conjugated with an enzyme (alkaline phosphatase, horse radish peroxydase . . . ). The immunoassay is based on the capture of the analyte of interest by the antibody capture grafted on the beads while the detection is performed using a secondary antibody targeting a different epitope. The analyte quantification is based on the amount of detectable secondary antibody. Using the magnetic droplet platform we developed, it is possible to incubate the beads with sample, to perform washing step, to incubate the beads with secondary antibody, to wash the beads and to perform the detection or to incubate the beads with an enzymatic substrate that will be transformed into a fluorescent product.

(79) Experimental

(80) Some more specific protocol details are now given.

(81) The droplet immunoassay was developed to quantify thyroid stimulating hormone (TSH) in sub 100 nL serum sample as biomarker for the neonatal diagnosis of congenital hypothyroidism. The magnetic beads and all the immunoassay reagents are from Immunometrics TSH kit except MUP from Sigma. The capture antibody and detection antibody are monoclonal targeting different TSH epitopes.

(82) The droplet immunoassay is performed as following: An 8 droplets train (100 nl each droplet) is generated containing all the reagents required to perform the TSH quantitation as depicted FIG. 4. Low volume droplets were generated using a robotized injector [2] and transported by a perfluorinated oil inside perfluoro-alkoxy capillaries (300 m ID, Sigma). Droplet actuation was ensured by syringes pumps (Nemesys, Cetoni). First the magnetic beads M (carboxylic acid, 1.05 m, MyOne Dynabeads, Invitrogen) are incubated with a horse serum spiked with various TSH concentrations ranging from 0 to 60 mIU/L. After an incubation time of 5 min, the immunological complex (capture antibody/TSH) 50 immobilized on the magnetic beads M is magnetically trapped by the magnetic tweezers 5 and 6. The beads are thus washed by flowing through the bead cluster 35 a TBS (tris buffered saline: 50 mM Tris.HCl, pH 7.4 and 150 mM NaCl) droplet to avoid the non-specific adsorption of proteins on the immunosupport. Using a second set of magnetic tweezers 7 and 8, the beads are magnetically confined and released in a droplet containing the secondary antibody coupled with the phosphatase alkaline enzyme (incubation time 5 min). Finally after a second washing step with 3 TBS droplets to remove unbound secondary antibody, a third set of magnetic tweezers 7a and 8a was used to release the particles in an enzymatic substrate (4-Methylumbelliferyl phosphate) droplet to perform the detection.

(83) The magnetic particles were manipulated all along the droplets train by transferring them from a drop to the next one using a set of micro machined tips (AFK 502, Imphy Alloys) with an apex curvature radius <50 m and magnetized on-demand with a home-made coil.

(84) Principle of Detection

(85) The principle of the detection consists in this example in using AP to transform a substrate into a product with properties detectably different. This could be, for instance, a change in colour, fluorescence, solubility, or redox properties.

(86) For instance, in this example antibody labelled with PA are used with MUP as substrate. As an important remark, for use of PA, all buffers used should be devoid of phosphate, in order to avoid the phosphate competing with the substrate. We used TBS buffer as previously mentioned. The fluorescence signal is monitored using an epifluorescence microscope with a high sensitivity camera and dichroic equipment. The filter used is a DAPI one, with (exc=358 nm et em=461 nm). The observed fluorescence, which signs the production of 4-methylumbelliferone, is directly measured in the chip in real time from the epifluorescence objective.

(87) Principle of Quantitation

(88) There are two main different formats to monitor ELISA reaction end-point ELISA and kinetic ELISA. Kinetic ELISA differs from end-point ELISA as it is based on the enzyme substrate reaction kinetics. Indeed when the substrate is present in great excess, there is a linear relationship between the enzyme concentration and the velocity of substrate turnover. Both ELISA format can be performed using such platform.

(89) This example reports TSH quantitation within droplet by kinetic ELISA. FIG. 6 reports the temporal variation of the fluorescence signal within droplets with various initial TSH concentrations. As expected from enzymatic reaction, for a given period of time, the fluorescence signal increases linearly with time whereas at later time the concentration of fluorescent product reaches a plateau whereas the slope of reaction curves increases with the TSH concentration.

(90) The immunoassay calibration curve was obtained by plotting the initial enzymatic rate against the TSH concentration (FIG. 5). The detection limit defined as three standard deviation of the background signal was 2 mIU/L equaling to 1.8 M (Planells 1975) which is comparable with conventional colorimetric kinetic ELISA (2.3 mIU/L). The complete immunoassay is performed in less than 10 minutes compared to the 2 h30 required for the conventional ELISA.

(91) As soon as the first step is performed on the drop train, it is possible to start a second analysis since tweezers are controlled independently. By this way, the immunoassay output is controlled by the droplet train length. This leads to an analysis throughput of 120 analyses per hour on sub 100 nL samples.

(92) Results and Discussion

(93) This strategy was applied to an immunoassay dedicated to neonatal diagnosis of congenital hypothyroidism (CH), a disease of high prevalence (1:2000 to 1:4000 newborns). The prognosis of infants treated early is excellent whereas untreated CH leads to severe development problems.

(94) The clinical diagnosis of CH is mainly based on an elevated concentration of thyroid stimulating hormone (TSH>30 mIU/L serum). Using this microfluidic platform, the whole immunoassay is performed in 10 min in sub-100 nL volume sample (as compared to 2 h30 and 200 L for conventional ELISA) which are essential criteria for neonatal diagnosis. The analytical sequence of the sandwich ELISA we developed is presented in FIG. 4. The quantitation was performed by monitoring continuously the increase of fluorescence as a function of substrate incubation time.

(95) The TSH concentration is thus determined by the slope of the reaction curve corresponding to the enzymatic initial rate (FIG. 6). The detection limit defined as three standard deviation of the background signal was 2 mIU/L equaling to 1.8 M (FIG. 5) which is comparable with conventional colorimetric ELISA, and meets the standard of congenital hypothyroidism diagnosis.

(96) FIG. 7 show that results in microfluidic droplets fulfill the standard sensitivity required for TSH detection (standard threshold 30 mIU/L in serum). Results (normalized) compared to conventional colorimetric batch analysis. Error bars refers to triplicate measurements.

(97) More generally, a sensitivity in the picomolar range paves the way for numerous analysis of disease biomarkers present in this concentration range in biological matrix. Current investigations on multiplexing show that, once started, the platform allows for continuous droplet-train generation and analysis associated to a maximum analysis rate above 120 analysis/hour and appears as very appealing for the development of a high throughput strategy.

(98) Conclusion

(99) We successfully developed a flexible immunoassay platform based on confined droplet and magnetic particle handling. The present work was applied to congenital hypothyroidism diagnosis but can be extended to almost any immunoassay. Our results show similar sensitivity performances compared to batch protocol while providing a 1000-fold volume reduction and a total analysis time shortening from 2 h30 to 10 min. This approach paves the way for automated and high throughput screening of biomarker on low volume sample.

Example 3: Magnetic Bead-Based Immuno-Agglutination Assay in Confined Droplets

(100) As shown in FIG. 9, an aggregation step is performed in water in oil droplets 30 and is induced by magnetic confinement to enhance magnetic beads (MBs) M collision frequencies thus favoring aggregates formation [17].

(101) Confined droplets 30 in fluorinated oil [18,19] allow individual compartmentalization preventing cross contaminations. Moreover, the possibility to generate them in large number in a pipeline format gives access to reliable and high throughput analyses in simple chip designs.

(102) First demonstration of the assay was performed (FIG. 9) using streptavidin coated MBs (1 m) M (surface-functionalized magnetic particles) and biotinylated phosphatase alkaline (b-PA) (target: 38) as a model. As detailed hereunder, the magnetic particles M provides a plurality of binding sites 39a.

(103) Droplets 30 were generated in a Teflon tube by sequentially aspirating defined volumes of oil and sample (containing the MBs M and the target 38, respectively) from a microtiter plate. (1) Droplets 30 were generated from a mixture of MBs M that was first incubated with b-PA 38 for 5 minutes and further transported in a Teflon tubing. The target 38 can thus be captured on the binding sites 39a of the particles M. The droplets containing free MBs were further transported towards magnetic tweezers (2) 5 and 6. When passing in between the magnetic tweezers 5 and 6, the MBs were magnetically confined to enhance aggregates 35 formation.

(104) Once passing the tweezers 5 and 6, the internal recirculation flows in the droplet induce shear forces that favors MBs re-dispersion. This process prevents non-specific aggregation but preserves specific interactions between particles, which stay in the aggregated state and still forms an aggregate 35a (FIGS. 9, 10A and B). As it can be seen from FIG. 9, the aggregated magnetic particles are linked to each other by a bridge 39 having the following structure: binding site 39a of a first magnetic particle-target 38-binding site 39b of a second magnetic particle different from the first.

(105) FIG. 10 shows bright field micrographs of MBs agglutination in droplets at 0 (A) and 100 ng/mL (B) of target. Changes in integrated transmitted light for the blank (S.sub.blank) and assay (S.sub.assay) experiments are monitored in transmission using simple visible low-cost USB camera 60. The signal is defined as (1-S.sub.assay/S.sub.blank). C) Calibration curves for b-PA target after 5 min incubation were obtained for three different concentrations of MBs: 1 (red), 2 (blue), 3 mg/mL (green). The insert is a focus on the 0-80 ng/mL target concentration range.

(106) The assembly comprises a light source 61 configured to irradiate the droplet, the optical detector 60 being configured to measure the quantity of light from the light source absorbed by the droplet.

(107) The light source 61 and the camera 60 allows to visualize the content of the detection zone 70 of the microchannel.

(108) The detection consists in measuring the change in the integrated light absorption across the droplet, induced by the aggregation process. Beads in the aggregated state occupy less area in the droplet perpendicular to the observation direction, and thus aggregation increases the transmission signal.

(109) When present in the detection zone 70, at least 10%, preferably 50%, preferably 70%, more preferably 80%, of the magnetic particles contained in the droplet may be in an aggregated form.

(110) FIG. 11 highlights that the application of B increases drastically the aggregation kinetics. Using a 1 mg/mL MBs concentration, the limit of detection was about 100 M (FIG. 10C) which meets the requirements of most immunodiagnostics.

(111) This droplet based platform provides analysis at high throughput (300 analyses per hour) and low cost. The set-up is composed of simple items while droplets format reduces the sample volumes down to 80 nL. Moreover, the integration of magnetic tweezers allows extraction and transfer of MBs from drop to drop thus allowing full automation of the assay steps.

(112) In a variant, the activable magnetic elements 5 or 6 may have the geometry depicted in FIGS. 12 to 14, and 15 to 15F.

(113) As shown in FIGS. 12 and 13, the activable magnetic element 5 comprises a tip 5a whose width w and thickness t decreases along the longitudinal axis Y of the tip 5. The activable magnetic element 5 shown has a convex shape.

(114) As shown in FIG. 15, the magnetic element 5 may have a parallelepiped shape with a tip having a decreasing section when moving towards the micro channel. The tip 5a extends as shown transversely, in particular perpendicularly, to the longitudinal X of the microchannel.

(115) In a variant, the magnetic element has a cubic shape.

(116) FIG. 16A further shows an embodiment of a directly activable magnetic element which is the core of an electromagnet.

(117) FIG. 16B shows an example of a soft magnetic element used as an activable magnetic element, wherein a permanent magnet can be brought in close proximity of said soft magnetic element by mechanical means.

(118) The permanent magnet can be rotated in order to have its pole in contact with one side of the core, in such case, the core is activated, or rotated in order to be no more in contact with such core, in such case the core is inactivated.

(119) FIG. 17, further shows an embodiment of a micro channel which comprises branching such as side branching 101, or cross-branchings 102. The branching areas are located along the micro channel in areas distant from the areas facing the activable magnetic elements 5 and 6 as shown.

REFERENCES

(120) [1] R. S. Sista, A. E. Eckardt, V. Srinivasan, M. G. Pollack, S. Palanki and V. K. Pamula, Heterogeneous immunoassays using magnetic beads on a digital microfluidic platform, Lab on a Chip, vol. 8(12), pp. 2188-96, (2008). [2] M. Chabert, K. D. Dorfman, P. de Cremoux, J. Roeraade and J-L. Viovy, Automated microdroplet platform for sample manipulation and polymerase chain reaction, Analytical Chemistry, vol. 78(22), pp. 7722, (2006). [3] H. Tsuchiya, M. Okochi, N. Nagao, M. Shikida and H. Honda, On-chip polymerase chain reaction microdevice employing a magnetic droplet-manipulation, Sensors and Actuators B, vol. 130(2), pp. 583-588, (2008). [4] D. Lombardi and P. S. Dittrich, Droplet microfluidics with magnetic beads: a new tool to investigate drug-protein interactions, Analytical and bioanalytical chemistry, vol. 399(1), pp. 347, (2010). [5] Clague, A., & Thomas, A. (2002). Neonatal biochemical screening for disease. Clinica Chimica Acta, 315(1-2), 99-110. [6] Shikida, Mitsuhiro, Takayanagi, Kentaro, Honda, Hiroyuki, Ito, H., & Sato, Kazuo. (2006a). Development of an enzymatic reaction device using magnetic bead-cluster handling. Journal of Micromechanics and Microengineering, 16(9), 1875-1883. doi: 10.1088/0960-1317/16/9/017. [7] Shikida, Mitsuhiro, Takayanagi, Kentaro, Honda, Hiroyuki, Ito, H., & Sato, Kazuo. (2006b). Development of an enzymatic reaction device using magnetic bead-cluster handling. Journal of Micromechanics and Microengineering, 16(9), 1875-1883. doi: 10.1088/0960-1317/16/9/017. [8] Zhang, Y., Park, S., Liu, K., Tsuan, J., Yang, S., & Wang, T.-H. (2010). A surface topography assisted droplet manipulation platform for biomarker detection and pathogen identification. Lab on a chip, (207890). doi: 10.1039/c01c00296h. [9] Long, Z., Shetty, A. M., Solomon, M. J., & Larson, R. G. (2009). Fundamentals of magnetactuated droplet manipulation on an open hydrophobic surface. Lab on a chip, 9(11), 1567-75. doi: 10.1039/b819818g. [10] Lehmann, U., Hadjidj, S., Parashar, V., Vandevyver, C., Rida, a, & Gijs, M. (2006). Two dimensional magnetic manipulation of microdroplets on a chip as a platform for bioanalytical applications. Sensors and Actuators B: Chemical, 117(2), 457-463. doi: 10.1016/j.snb.2005.12.053. [11] Lafranchi, S. H. (2010). Newborn screening strategies for congenital hypothyroidism: an update. Journal of Inherited Metabolic Disease, 33, 225-233. doi: 10.1007/s10545-010-9062-1. [12] Al Hetlani et al., MicoTAS 2010 proceedings, CBMS publ, pp 1817-1819. [13] Dorfman, K. D., Chabert, M., Codarbox, J.-H., Rousseau, G., Cremoux, P. de, & Viovy, J.-L. (2005). Contamination-free continuous flow microfluidic polymerase chain reaction for quantitative and clinical applications. Analytical chemistry, 77(11), 3700-4. doi: 10.1021/ac050031i. [14] Song, H., Tice, J. D., & Ismagilov, R. F. (2003). A microfluidic system for controlling reaction networks in time. Angewandte Chemie, 42(7), 767-772. [15] A One-step homogeneous immunoassay for cancer biomarker detection using gold nanoparticle probes coupled with dynamic light scattering, X. Liu, Q. Dai, L. Austin, J. Coutts, G. Knowles, J. Zou, H. Chen and Q. Huo, Journal of the American Chemical Society, 130, 2780 (2008). [16] Three-Dimensional Magnetic Focusing of Superparamagnetic Beads for On-Chip Agglutination Assays R. Afshar, Y. Moser, T. Lehnert and M. A. M. Gijs, Analytical Chemistry, 83, 1022 (2011). [17] Acceleration of the recognition rate between grafted ligands and receptors with magnetic forces, J. Baudry, C. Rouzeau, C. Goubault, C. Robic, L. Cohen-Tannoudji, A. Koenig, E. Bertrand and J. Bibette, Proceedings of the National Academy of Sciences of the United States of America, 103, 16076 (2006). [18] Reactions in Droplets in Microfluidic Channels, H. Song, D. L. Chen and R. F. Ismagilov, Angewandte Chemie International Edition, 45, 7336 (2006). [19] ABO, D Blood Typing and Subtyping Using Plug-Based Microfluidics, T. R. Kline, M. K. Runyon, M. Pothiawala, and R. F. Ismagilov, Analytical Chemistry, 80, 6190 (2008).

(121) The expression comprising/containing must be understood, unless otherwise specified, as comprising at least one/containing at least one.

(122) The expression forming a must be understood, unless otherwise specified, as forming at least one.

(123) The expression comprised between . . . and . . . must be understood, unless otherwise specified, as including the bounds.