GB2510653A - Detecting low-abundant analyte in microdroplets - Google Patents

Abstract

A method to generate aqueous droplets (femtodroplets) in oil wherein the droplets have a volume of less than 500fL and are generated at a rate of at least 50000 droplets per second, comprises injecting a dispersion phase into a continuous phase flowing in a microchannel such that the two phases cross each other resulting in the formation of droplets of the dispersion in the continuous phase wherein the microchannel has a constriction or narrows in the region where the phases cross each other. A method of storing femtodroplets using a microfluidic storage component which has a microfabricated elastomeric structure and contains an elastomeric block wherein the block is deflectable into a microchannel to stop or trap femtodroplets. A method of analyzing the concentration of a molecule using femtodroplets comprises mixing the molecule of interest with capture particles and immobilizing the molecule, encapsulating the capture particle which contains the molecule in a femtodroplet, storing the femtodroplet in a microfluidic device, interrogating the femtodroplets and determining the concentration of the molecule based upon the results. Applications include the immunoassay, digital PCR, individual molecule resolutions, nanoparticles synthesis, biodegradable polymer particle production and single molecule analysis.

Description

TITLE OF INVENTION:

Detecting low-abundant analyte in microdroplets r o c'J

FIELD OF INVENTION:

The present invention relates to systems and methodsfor detecting analyte molecules or particles in a fluid sample and in some cases, determining a measure of the concentration of the molecules or particles in the fluid sample. Methods of the present invention may comprise immobilizing a plurality of analyte molecules or particles with respect to a plurality of capture particles. At least a portion of the plurality of capture particles maybe spatially separated into a plurality of locations. A measure of the concentration of analyte molecules in a fluid sample may be determined, at least in part, on the number of reaction vessels comprising an analyte molecule immobilized with respect to a capture particle. In some cases, the assay may additionally comprise steps including binding ligands, precursor labeling agents, andlor enzymatic components.

This invention relates to methods for microfluidic generation and storage of droplets, for fabrication of microfluidic devices. Embodiments of the methods are particularly useful for single-molecule counting immunoassay and polymer particle synthesis. -

BACKGROUND OF THE INVENTION:

Water-in-oil droplets are emerging as a potentially powerful technology to quantitatively study compartmentalized reactions of single enzyme molecules or single cells because the concentration of reactiOn products or secreted molecules exceed the detection threshold much more rapidly in small confined volumes than in bulk solution. Iti order for the enzymatic product to be detectable using epifluorescence microscopy, the volume of the reaction chamber containing the enzyme and its fluorogénic substrate have been reduced to less than 100 femtoliter. In this volume, a single molecqle of enzyme has a concentration of-' 17 picomolar, enabling substrate turnover to dominate 0) proees'ses such as uncatalyzed hydrolysis, which in turn allows rapid accumulation and detection of i-the product. Due to their inherent scalability, droplet-based platforms could enable numerous single-molecule assays to be performed in parallel.

O According toliterature written by Rotman et al [B. Rotman, Proc. NatI, Acad. Sci. U. S. A. 1961, 47, 1981] and Lee et al [A. 1. Lee, J. P. Brody, Biophys. J. 2005, 88, 43031, ultra-small droplet with * volumes ranging from 0.5 fL to 2 pL have been used to detect the activity of single enzyme (\J molecules, but the polydispersity of the emulsions used limited the precision and throughput of these studies.

According to literature written by Theberge et al [4. B. Theberge, F. Courtois, Y. Schaerli, M. Fischlechner, C. Ahell, F. Rollfelder, W. T. S. Huck, Angew. Chem., lnt. Ed. Engl. 2009, 49, 5846], Chiu et al [D. T. Chiu, R. M. Lorenz, 0. D. M. Jeffries, Anal. Chem. 2009, 81, 5111], Quo et al [M.

T. Guo, A. Rotem, J. A. Heyman, D. A. Weitz, Lab Chip 2012], there has been tremendous progress in the development of miorofluidics-based droplet platforms for the on-chip formation and manipulation of monodisperse droplets, and the associated use of a range of fluorescence-based techniques for high-throughput and highly sensitive analysis of droplet contents. Existing microfluidic devices generate highly monodisperse droplets at the pico-to nanoliter scale. In such volumes, according to literature written by Joensson et al [H. N. Joensson, M. L. Samuels, F. R. Brouzes, M. Medkova, M. IJhlen, D. R. Link, H. Andersson-Svahn, Angew. . Chem., mt. Ed. Engl.

2009, 48, 2518.], several hours of enzymatic activity are required to turn over sufficient substrate for single enzyme molecule detection. Furthermore, maximal droplet generation rates are in the 10 kHz range, limiting high-throughput measurements of fast reactions.

The gold standard immunoassay, ELISA (enzyme-linked immunosorbent assay), enables the detection of biomarkcrs at concentrations above pieomolar (1012 M), but there remains an unmet clinical need for detection of biomarkers of neurodegenerative diseases and cancers that are present in biological fluids at concentrations in the range of l012 _10.16 M;the ability to detect single enzyme molecules provides a means to quantitate such low abundance markers.

According to Literature written by Rissin et al [D. M. Rissin, C. W. Kan, T. G. Campbell, S. C Howes, D. R. Fournier, L. Song, T. Piech, P. P. Patel, L. Chang, A. J. Rivnak, E. P. Ferrell, J. D. Randall, G. K. Provuncher, D. R. Walt, D. C. Duffy, Nat. Biotechnol. 2010, 28, 595.], Zhang ci al [H. B. Zhang, S. Nie, C. M. Etson, R. M. Wang, D. R. Walt, Lab Chip 2012, 12, 2229.], Kan et al [C. Vt. Kan, A. J. Rivnak, T. 0. Campbell, T. Piech, D. M. Rissin, M. MosI, A. Peterca, H. P. Niederberger, K. A. Minnehan, P. P. Patel, E. P. Ferrell, R. E. Meyer, L. Chang, D. H. Wilson, D. R. Foumier, D.C. Duffy, Lab Chip 2012, 12, 977.] and Kimet al [S. H. Kim, S. Iwai, S. Araki, S. Sakakihara, R. lino, I-I. Noji, Lab Chip 2012.], one promising approach uses the turnover of a fluorogenic substrate by single enzyme molecules within well-arrays as the basis for ultra sensitive digital ELISA.

However, the need for mechanical fabrication of these femtôliter reaction chambers places inherent limits on the scalability and flexibility of ultra sensitive diagnostic assays, which could be overcome using a droplet-based approach.

SUMMARY OF INVENTION: -

According to the present invention there is therefore provided a metho4 of fabricating a multilayered microfluidic device that enables the generation and on-chip manipulation of highly * monodisperse emtoliter droplets at frequencies up to a few mega-hertz. This innovation allows the * measurement of enzymatic activity of single enzyme molecules in a few minutes, a property that have been exploited to construct a bead-based ELISA for the detection of a low-abundance protein biomarker. -C') -We invented a flow focusing nozzle having locally shallower depth and width to obtain a substantial enhancement of flow speed without a significant increase of the internal pressure. The CO local constriction is introduced within a section of the device, where the channel dimensions are 0 reduced (Figure la-b).

C\1 We invented a microfluidic component for storing femtodroplets for a sufficient time to monitor C'IJ chemical reactions therein. A wide and shallow storage area is integrated in the microfluidic device to trap and keep femtodroplets for long duration of time enough to accumulate certain amount of products. The storage area is divided into a few tens or hundreds of traps, each of which is isolated by monolithic microfluidic valves (Figure lb-d).

We invented a method to measure the enzymatic activity of individual enzyme molecules using the fentodroplets in the microfluidic device. The enzymatic activity of individual molecules can be interrogated in femtodroplets. Various concentrations of enzyme were encapsulated in femtodroplets with a fluorogenic substrate. As the enzymatic turn-over starts at the droplet generation, the initiation of chemical reaction in stored femtodroplets is perfectly synchronized, and thus can be. precisely monitored in time. The time course fluorescence of femtodroplets stored in.

each trap is imaged in order to yield kinetic information of the chemical reactions in each droplet (Figure 2).

We invented a digital immunoassay using the femtodroplet assay and a bead-based antibody binding assay, termed the femtodroplet immunoassay, which is able to quantify very low concentration of biomarkers. We exploited the ability of the femtodroplet assay to detect the presence of single enzymes in order to measure concentrations of target analyte which is conjugated with enzyme reporters (Figure 3, Figure 4).

We invented a method to perform identical repetitive ferntodroplet immunoassay in a single assay.

The embedded microfluidic valve is conveniently controlled to flush stored femtodroplets out of and reload freshly generated femtodroplets into those traps by application and release of external pressure. This is done in seconds due to the extremely frequent droplet generation so that it enables us to conduct identical repetitive assays in every a few minutes for demanded time duration.

We iivented a method to identify presence of beads using fluorescence of protein. We found that the capture-antiboby conjugated beads are fluorescent due to the intrinsic fluorescence of immunoglobin. The bead fluorescence is strong enough to be observable in red-fluorescence and at a same time weak enough for single enzyme activity in the femfodroplet to be differentiated in green-fluorescence so that it enables us to count the number of beads more accurately and comfortably than when using the bright field images (Figure 4a).

We invented a method to measure the molecular weight of target analyte. We estimated the molecular weight of target proteins with the femtodroplet immunoassay. As Figure 3d shows the relation between mass concentration (pg/mI) and molar concentration (pM) obtained from a sample preparation and the ferntodroplet immunoassay respectively, it is possible to cakulãle Mw.

We invented a microfluidic device to store and interrogate more than millions of femtodroplets in a single assay. The device has at least 100 traps in order to simultaneously incubate and monitor more than millions femtodroplcts. The storage is serially extended to be able to locate the nuthbers of traps in a device (Figure 5a). Also, a device having multiple nozzles can be used. Each nozzle is parallel connected with its own storage that is designed to have a few tens of traps to minimally enhance the resistance.

tY) We invented a method to measure the frequency of femtodroplet generation in-line. We employed -.-the frequency resonance impedance spectroscop' that is integrated in the device. It measures the impedance change due to the water droplets passing through an aperture placed between two CO electrodes, which is fabricated in the mierofluidic device (Figure Sb). The real-time measuremçnt of C the impedance simultaneously determines the frequency and volume of femtodroplets sothat the target volume can be precisely obtained by controlling the flow rate of fluids.

C'SJ We invented a method to enhance the detection throughput of the femtodroplet immunoassay by encapsulation of multiple beads in a droplet. In order to encapsulate one bead per droplet only 10% of droplets are occupied by beads and the rest, 90%, have no bead. To get rid of this inefficiency of droplet usage multiple beads in a droplet can be encapsulated. Encapsulation of multiple beads maximizes the usage of droplets, thus reduces the time to detect the target molecule and speeds up the throughput; therefore it enhances the sensitivity.

BRIEF DESCRIPTION OF DRAWINGS:

Figure 1. Features of the microfluidic device used for fcmtodroplet generation and manipulation.

(a) Femtodroplet formation at the nozzle of the microfluidic device. Droplets with a typical volume of 32 if. are generated at a frequency of around 3.5 io per second. (b) Photographofthe whole multilayered PDMS device. The upper layer consists of the nozzle (10 jim wide x 5 m deep), flow channels (100 j.tm wide x 25 jim deep) and storage compartments (2 miii wide x 7mm long > 5 jim deep), with a capacity for -2 x fl35 femtodroplets. The bottom layer houses the monolithic valves used to control droplet flow and isolate the fraps. There are injection holes for introduction of fluids into the device; the outer two for oil and the others for aqueous solutions. When the main valve is closed, the stream of femtodroplets is directed into the storage region (stream path 1). lf this valve is opened, droplets flow out of the device by stream path 2 due to the lower flow resistanëe encountered. (c) Image of traps used for femtodroplet storage and isolation The contents of the * traps are manipulated by the action of networks of embedded monolithic valves in response to external pressure. (d) Vertical schematic of the storage structure. When pressure is applied, the thin PDMS membrane (15 m thick) bends up to seal off the flow and traps the femtodroplets. (e) Generation frequency and volume of femtodroplets as a function of theoil flow rates (Q0i) at a constant water flow rate (Qwater) of 40 p1/hr. In order to generate 32 ft. droplets at a frequency of 350 kHz, 230 jiL/hr of oil and 40 p1/hr of water in flow rates were introduced. The dashed line is a prediction curve, fitting the experimental data. The linearity (r-square) is 0.9. (0 The histogram of the droplet volume. The volume shown is 35.5 fL and the standard deviation is 0.66 ft at 40 p.L/hr of water and 220 RL/hr of oil.

Figure 2. Measurement of the activity of single 3-gaIactosidase molecules in femtodroplets.

(a) Femtodroplets stored in a trap. The supporting posts maintain the shallow trap structure (5 jim deep). Droplets were packed in a monolayer in a trap (right). (b) Images showing green fluorescence resulting from hydrolysis of FDG (250 pM) by 3-galactosidase (1.5 x io unit/mi, equivalent to 2.1 pM) in femtodroplets after 1 and 10 mm. The bright spots represent femtodroplets.

enclosing a single enzyme molecule, in which fluorescein reaction product is generated. (c) Representative time traces of enzyme activity measured in femtodroplets that contain either one 3-galactosidase molecule or none. The fluorescence was measured every minute and converted to concentrations of fluorescein uing a calibration curve after correcting forphotobleaching. The black dashed line represents a threshold, definedabove three standard deviations of the background at 10 minute. The positive traces show a range of activities. (d) Product concentration increase per single copy of 3-galactosidase as a function of substrate (500, 250; 125, 63, 25, 13 pM FDG).

Taking the time derivative of the fluorescein concentration produced in the en.zymatically-active droplets in Fig. 2c yields the concentration increase. The error bar is the standard deviation. The data can be fit to the Michaelis-Menten equation, v = [S]/(K + [SI), giving a good correlation Ct) (r2 = 0.97) and a K,, of 90 ± 26 pM. (e) Fluorescence micrographs of traps after 10 mm incubation T at various enzyme concentrations. The fraction of stored ferntodroplets that show product formation Cy) varies in a concentration-dependent manner. (0 Plot of the prepared concentration (where one unit O of enzyme hydrolyzes 1 pM of o-nitrophenyl f3-D-galactoside to o-nitrophenol and D-galactose per minute at pH 7.3 at 37 °C).vs. the experimentally-determined molar concentration of - ç galactosidase. The dotted line represents a linear fit.

Figure 3. Schematic of single molecule immunoassay using femtodroplets.

(a) Binding of an antigen to antibody-coated beads; a single bead-captured target molecule. is subsequently sandwiched by a biotinylated detection-antibody and a streptaidin-3-galactosidase conjugate. (b) Beads with or without an immunocomplex are singly encapsulatód in femtodropiets with a substrate (FDG) and subsequently in-line incubated in the trap to accumulate the fluorescent product of single enzyme reporter. Each trap can enclose about 5 x l0 droplets so the capacity of the current storage is about 2 x i05 droplets. The stored droplets can be completely flushed out and reloaded in 10 seconds due to the extremely high-speed generation of droplets. (c) After the in-line incubation, three populations of femtodroplets are observed, i) droplets containing no bead ii) those containing a bead without irmnunocomplexes and iii) those containing a bead with an immunocomplex exhibiting a positive fluorescence signal due to the enzymatic activity of single enzyme reporter. The numerical ratio of (iii) to ((ii) + (iii)) yields the concentration of the targct molecules. Thus, the larger the number of available droplets in a measurement is, the lower the detection sensitivity can be accomplished in a given assay time.

Figure 4. Detection of PSA using femtodroplet assay.

(a) Brightfield, red-and green-fluorescence images (left to right) of stored femtodroplets containing anti-PSA coated beads (10 pM) and substrate (FDG, 250 pM) after a 10-minute incubation period following immunoassay (in the presence of a 140 pg/mI. concentration of PSA) and subsequent encapsulation. Bright spots in the red-fluorescence micrograph result from beads conjugated with the capture-antibody, while those in the green-fluorescence image are due to femtodroplets in which enzymatic activity has occurred. The green circles in all images represent droplets that contain a bead and show enzymatic activity, while the red circles in the bottom left corner of each panel indicate a single femtodroplet that exhibits enzymatic activity but which does not contain a bead, indicating the presence of unbound enzyme in the droplet. (b) Plot of the molarity of PSA measured by the droplet-based immunoassay vs. the prepared concentration. Molar concentrations were calculated from a Poisson distribution function as described in Fig. 2f'except that the fraction of beads encapsulated in droplets showing enzymatic activity (e.g. the number ratio of green circles to fluorescent beads in Fig. 4a) and the known bead concentration (10 pM) were used instead of the inactive fraction and volume of the femotodroplets. The doffed line represents a linear fit Figure 5. (a) A next generation microfluidic device having more than 1100 traps. (b) Schematics of the impedance spectroscopy

DETAILED DESCRIPTION OF PREFERRED EMBODIMENT:

We will describe a microfluidic device that is able to generate and manipulate droplets with volumes of I -100 if. at MHz frequencies. This femtoliter microfluidic droplet-based approach enables the measurement of the activity of a single copy of am enzyme and can be exploited to quantify very low-abundance biomarkers by integrating a bead-based immunoassay with direct counting of individual enzyme molecules for developing a highly sensitive diagnostic test, The fluidic femtodroplet reaction chambers used in this study offer significant advantages due to the robustness and flexibility of the microfluidic circuit compare to the digital ELISAs reported by Rissin et al[Nat. BiorechnoL 28, 595-U525 (2010)]: extremely high-speed generation and manipulation of fast-flowing droplets, the ability to carry out replicate assays without replacing (1') hardware enabling a significant enhancement of the sampling size, ease of automation and .-. integration with other fluidic sample preparation modules and the possibility of varying the size of the reactors at will. C')

0 1. Generation and manipulation of femtoliters volume microfluidie droplets We have developed a microfluidic device that is able to generate controllably and manipulate water droplets in oil of 1 -100 ferntoliter volume -which we call femtpdroplets'-at frequencies >1 MHz C\J (Figure 1 a, I e). Microfluidic droplets can be generated by shearing one fluid (water) by a second immiscible one (oil). In order to i3roduce small.water droplets at high frequencies, a large shear force and low intel-facial tension at the oil-water interface are required according to Yobas et al[Yobas, L., Martens, S., Ong, W.L. Ranganathan, N,Lab Chip 6, 1073-1079 (2006)]. Large shear forces can be generated by either applying a higher ficiw rate of oil or reducing the channel dimensions in order to increase the flow speed. However, high flow rates can lead to difficulties in device operation and smaller channel dimensions produce high internal pressure, inversely proportional to the fourth power of the channel diameter according to Beebe et al[Beebe, D.J., Mensing, GA. & Walker, G.M. Annu. Rev Bionied. Eng. 4, 261-286 (2002)]. In order to substantially enhance the flow speed during droplet formation without generating high internal pressure throughout the flow channel in the device, a flow-focusing nozzle was integrated into the design of our device. This strategy introduces a local constriction within a local section of the device, for example, 300-jim; where the channel dimensions are reduced, for example, from.100 jim (width depth) to 10 5 jim (Figure la-b). This nozzle enables the controlled generation of highly monodisperse aqueous droplets in oil, for example, fluorinated oil (HFE-7500, NovecTM, 3M), previously mixed with a surfactant, for example 5% w/w, at frequencies of lO -106 Hz (Figure le-f). The interfacial tension(IFT) between the oil and water exhibited by this mixture is extremely low, for example -3 mN/m, which allows the generation of small droplets at much higher frequencies than is possible with other oils, e.g. silicone oil (IFT-38 mN/m) and mineral oil (IFT-5 I miN/m). The frequency of droplet-formation was measured using a confocal optical setup, and the droplet volume calculated from the formation frequency and the flow rate of water.. Using the cia-rent experimental setup, the frequency is maximally measurable up to 1.3 MHz, leading to a femtodroplet volume of 8.6 fL. However, very stable droplet generation at an oil flow rate of 480 tL/hr was observed, where the droplet-generation frequency is expected to be 3.1 MHz according to the curve fit, implying a femtodroplet volume of 3.6 IL. This droplet generation frequency is about two orders of magnitude faster than previously reported according to Theberge,A.B. et al.{Aiigew. C/win., In!. Ed. Engi. 49, 5846-5868 (2009)1. We conclude that the low interfacial tension, for example less than 10 miNim and the locally narrow flow-focusing nozzle design, for example a local section of less than 10 j.tm depth and about 300 rn lông, are key features enabling controllable generations of £emtoliter droplets at millions-hertz frequencies. The femtodroplets formed using our device provide discrete reaction compartments that are small enough to enable the prducts of one molecule of enzyme to be detected within minutes by epifluoreseence microscopy but also large enough to be manipulated fluidically.

Once single enzyme molecules and the fluorogenic substrate have been encapsulated, it takes a few minutes to accumulate a measurable amount of fluorescent product. A storage area, for example 2 mm 7 mm Mm (length width x depth), was therefore integrated into the mierofluidic device to store femtodroplets while the enzymatic reaction occurs (Figure Ib). The storage area is divided into a number of traps, for example 40 traps with for example 300 Mm x 300 Mm wide, isolated by monolithic microflüidic valves (Figure lb-d). As the depth, for example S m, of the storage area is comparable to the diameter of the femtodroplets, droplets stored in the microfluidic device are packed into a monolayer that allows fluorescence measurements of individual droplets using a simple epifluorescence microscope (Figure 2a). Trapping the fcmtodroplets in this way allows enzymatic activity of specific enzymes to be monitored continuously inside thousands of droplets simultaneously (Figure 2b). An embedded microfluidic valve is used to flush stored droplets out of the traps and reload freshly-generated femtodroplets by application and release of external pressure, CO for example about 50 psi. This process takes only about 10 seconds due to the extremely high f-frequency of droplet generation and it is therefore not rate-limiting for assay repetition.

cv) 2. Measurement of enzymatic reaction of individual enzyme molecules 0 We first determined the time required for individual molecules of f3-galactosidase encapsulated in 32 fL droplets to generate sufficient fluorescence signal to be detectable above the background from 250 tM of a substrate (fluorescein-di--D-galactopyranoSide, FDG). As enzymatic turnover starts at droplet generation, the initiation of the chemical reaction in each femtodroplet occurs within a second of each other, and so can be precisely monitored temporally. The time course of fluorescence generation in approximately 5 x io3 femtodroplets stored in each trap was imaged at enzyme concentrations of up to 3 x jØ2 unit/mL (equivalent to about 40 pM) where likelihood of enzyme occupancy of each droplet is c 0.8 (Figure 2e). After incubation for 10 minutes, two populations of droplets were clearly visible (Figure 2b). The fraction of bright femtodroplets (Figure 2e) -withintensities separated from the mean fluorescence of the other dark droplet population by> 3 s.d. (Figure2c) -followed a Poisson distribution as a function of prepared enzynie concentration, as expected if the observed product formation is due to the activity of single molecule of -galactosidase. The fraction of enzymatically-inactive femtodroplets (i.e. ii = 0) was inserted into a Poisson distribution function, JO) .e'/n!, where n describes the number of enzyme molecules in a droplet, yielding the average occupancy per droplet (A). The molar concentration of enzyme was then calculated from the average occupancy and the femtodroplet volume (32 if.). The linear relation between the prepared cOncentration and the determined concentration of 3-galactosidase in Figure 2f confirmed that the enzymatic activity observed in the bright femtodroplets is due to single enzymemolecules.

The enzymatic activity of individual molecules of 3-galactosidase (3.8 x 1 o-3 unit/mL, equivalent to about 5 pM) was also kinetically-characterized in femtodroplets at various substrate concentrations with each experiment monitoring more than 150 enzyme moledules stored in each trap (Figure 2c-d).

The lowest substrate concentration was 13 MM, so enough substrate is present to eliminate the effect of substrate depletion. The averaged enzymatic activity of individual enzyme molecules depends asymptotically on substrate concentrations according to the Michaelis-Menten equation (Figure 2d).

The averaged Michaelis constant (1(m) of femtodroplet-encapsulated -ga1aetosidase was 90 RM, which closely matched that measured in bulk (124 tiM). However, single-molecule measurement of enzyme kinetics revealed significant molecule-to-molecule variation in activity: the coefficients of variation (ratio of the standard deviation to the mean) are 0.64 and 0.13 for single enzyme and ensemble measurements, respectively, ibis wide distribution likely reflects the existence of considerable variation of activities within a population of enzyme, which has also been reported by other laboratories.

3. Detection of a cancer biomarker using a femtodroplet assay The ability to sensitively detect -galactosidase, a typical reporter ehzyme, paves the way for ultrasensitive diagnostics usinE a bead-based ELISA to quanti' very low coilcentrations of the biomarker prostate-specific antigen (PSA) reported by a single enzyme. A mcnoclonal antibody to the target protein was covalently coupled to polystyrene beads, for example 1 jtm diameter, to enable capture in PBS bufftr and subsequent detection of PSA in a sandwich complex containing a detector antibody specifically bound to a 3-galactosidase reporter (Figure 3a). The capture antibody-functionalised beads exhibited red autofluorescence, possibly due to the intrinsic fluorescence of immunoglobin according to Eftink, M.R.Methods Biochern. Anal. 35, 127-205 (1991)]. This made it possible to count the number of beads by fluorescence imaging more easily than by using brightfield illumination, without interfering with the detection of enzymatically-produced fluorescein in the green part of the spectrum (Figure 4a).

At the end of each experiment three different populations of femtodroplets were observed: i) * droplets containing no bead; ii) droplets encapsulating a bead but without detectable enzymatic activity arid iii) droplets containing a bead and a positive signal in green-fluorescence microscopy, C') corresponding to the presence of active enzyme conjugated to the target protein (Figure 3c). Since r-the concentration of PSA was lower than the bead concentration during anchoring of the target protein to the beads, Poisson statistics dictate that most beads capture either a single enzyme CO reporter or none. As the bead concentration was known, the fraction of bead-containing C femtodroplets that exhibit enzymatic turnover to the total number of beads was used to calculate the concentration of PSA (Figure 4b). The linear relationship obtained between the know" mass concentration and the experimentally-determined molar concentration confirmed that this a3proach C\IJ can be used to quantify a low-abundance biomarker. Since the molar concentration in commercial PSA preparations is not known, the accuracy and precision of the assay was verified by comparing the molecular weight of PSA calculated from the experimental data (36.9 ± 1.1 Wa) to the literature value (36 kDa). In the negative control -where the assay conditions were identical except that PSA was omitted -over 3,700 femtodroplets containing capture beads were analyzed, none of which exhibited detectable reporter fluorescence after incubation.

Another source of false positive signal would be free enzyme, not bound to.beads. However, as femtodroplcts enclosing a bead were specifically identified by their red fluorescence, those false positive signals were easily ruled out (Figure 4a). As a result, the lowest detectable analyte concentration was ultimately determined by the capacity of the current femtodroplet traps. As around 1,900 droplets encapsulating beads were analyzed per measurement, the theoretical limit of detection (i.e. the concentration required to generate an average of one fluorescent droplet in each experiment) is 5 flvl.