Subscriber access provided by University of Texas Libraries

Langmuir is published by the American Chemical Society. 1155 Sixteenth Street N.W.,Washington, DC 20036Published by American Chemical Society. Copyright © American Chemical Society.However, no copyright claim is made to original U.S. Government works, or worksproduced by employees of any Commonwealth realm Crown government in the courseof their duties.

Letter

Sugar Additives Improve Signal Fidelity for ImplementingTwo-phase Resorufin-based Enzyme Immunoassays

Patrick A. Sandoz, Aram J Chung, Westbrook M. Weaver, and Dino Di CarloLangmuir, Just Accepted Manuscript • DOI: 10.1021/la5004484 • Publication Date (Web): 28 May 2014

Downloaded from http://pubs.acs.org on June 3, 2014

Just Accepted

“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are postedonline prior to technical editing, formatting for publication and author proofing. The American ChemicalSociety provides “Just Accepted” as a free service to the research community to expedite thedissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscriptsappear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have beenfully peer reviewed, but should not be considered the official version of record. They are accessible to allreaders and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offeredto authors. Therefore, the “Just Accepted” Web site may not include all articles that will be publishedin the journal. After a manuscript is technically edited and formatted, it will be removed from the “JustAccepted” Web site and published as an ASAP article. Note that technical editing may introduce minorchanges to the manuscript text and/or graphics which could affect content, and all legal disclaimersand ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errorsor consequences arising from the use of information contained in these “Just Accepted” manuscripts.

1

Sugar Additives Improve Signal Fidelity for

Implementing Two-phase Resorufin-based

Enzyme Immunoassays

Patrick A. Sandoz†, Aram J. Chung††, Westbrook M. Weaver, Dino Di Carlo

*

Department of Bioengineering, University of California, Los Angeles, 90095, CA, USA.

KEYWORDS: Multiphase microfluidics, Water-in-oil emulsion, Enzyme immunoassay,

Resorufin, Fluorophore stability.

Page 1 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

2

ABSTRACT

Enzymatic signal amplification based on fluorogenic substrates is commonly used for

immunoassays, however, when transitioning these assays to a digital format in water-in-mineral

oil emulsions such amplification methods have been limited by the leakage of small reporting

fluorescent probes. In the present study we used a microfluidic system to study leakage from

aqueous droplets in a controlled manner and confirmed that the leakage of fluorescent resorufin

derivatives is mostly due to the presence of the lipophilic surfactant Span80, which is commonly

used to preserve emulsion stability. This leakage can be overcome by the addition of specific

sugars that most strongly interfered with the surfactants ability to form micelles in water.

Application of the microfluidic system for quantitative analysis of droplets as well as

implementation of the described sugar additives would allow for alternatives to fluorinated

surfactant-based platforms and improve the signal fidelity in enzyme immunoassays

implemented through multiphase microfluidics.

Page 2 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

3

Introduction

Over the last decade, a variety of multiphase platforms have emerged in microfluidic formats to

perform analysis of single cells or molecules (1-3). Miniaturization has allowed confinement of

small and uniform volumes and enabled the emergence of digital sensing in which the sample is

fractionated to sufficiently small volumes so that one or zero target molecules or cells is

statistically confined to each volume. The distribution of volumes with a single object

encapsulated then follows Poisson statistics and the sample concentration can be calculated (4).

The most widely demonstrated digital platforms use water-in-oil confinement (5, 6) or soda-lime

glass chambers (7) and quantify the concentration of specific nucleic acid sequences which

allows the determination of viral loads or the diagnosis of rare allele presence (8, 9). Various

other platforms demonstrate particle polymerization (10), single-cell or bacteria analysis (11) or

quantification of molecular kinetics (12).

The majority of digital platforms perform single-molecule detection using the polymerase chain

reaction (13) or enzyme immunoassays (14) within confined volumes and offer highly sensitive

quantification at the attomolar scale. During confined PCR reactions intercalating fluorescent

dyes that have increased quantum yield when associated with double stranded DNA are used for

readout. For enzyme-amplified immunoassays a secondary antibody conjugated to an enzyme is

often used to report the presence of the antigen by converting fluorogenic dyes such as resorufin

or fluorescein precursors. A common limitation of these platforms remains the ability to integrate

millions of stable, discretized and confined volumes in a restricted space for imaging and

readout. As microfluidic networks have a minimal required fabrication resolution only a

restricted number of rigid confinements can be put into a platform. Conversely, droplet based

Page 3 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

4

microfluidic systems offer continuous encapsulation of picoliter aqueous systems into a

surrounding oil phase. This approach is not limited by microfluidic fabrication and can produce

thousands of droplets per second. However, a significant challenge in creating an effective and

robust droplet system is maintaining both the physical stability of the droplet and the signal

fidelity within them. The stabilization of the phases is usually improved by the addition of a

surfactant which presents two moieties (one hydrophobic, the other hydrophilic) and reduces the

free energy of the two-phase separation. However, the presence of surfactant leads to more

complex interactions between the different phases and leakage of dyes from the droplets to the

outer phase. As a consequence previous platforms were developed on a case-by-case basis

depending on the chemistry of the system (15).

Fluorosurfactants are commonly used in these assays because of reduced dye leakage and

oxygen permeability properties; however, there are disadvantages which motivate use of non-

fluorinated systems. Isothermal droplet-based DNA amplification as well as living cell assays in

droplets make use of oxygen permeable fluorinated oil and fluorosurfactant (16, 17).

Thermocycled droplet digital PCR approaches have in some cases made use of the increased heat

transport and thermal stability of silicone-based oil and surfactants instead (5) or also employed

fluorinated oil and fluorosurfactant (13). Both combinations allowed confinement stability.

However, fluorinated surfactants have several drawbacks. First of all, their synthesis is a

complex process and the number of commercially available and affordable products remains

restricted. Even more important, these fluorinated surfactants, in particular perfluorooctanoic

acid (PFOA) and perfluorooctane sulphonate (PFOS), are harmful and due to their persistent

lifetime, they can accumulate in the blood of many species including humans. Their level of

Page 4 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

5

toxicity has even forced the U.S. Environmental Protection Agency to launch a program for

decreasing their industrial production in 2006 (18). As a consequence, this has led researchers to

turn to alternative solutions. One cost-effective non-toxic and stable two-phase system combines

for example mineral oil with a mix of sorbitan-oleate (Span80 <Sigma>) and polysorbate

(Tween80 <ThermoFisher>) as surfactant (15). This system resolves the problem mentioned

above for fluorinated oil and surfactant.

However, a general issue of stability for small molecules like fluorophores in the aqueous phase

of an emulsion remains. The low stability of amphipathic readout fluorophores in water-in-oil

emulsions has been reported several times (19-22). Recently, two groups brought new insights

characterizing the small probe leakage in water-in-oil emulsions. Chen et al. reported a model for

the different factors contributing to the leakage of small fluorescent probes out of a droplet (23).

The authors characterized the crosstalk among neighboring droplets from the droplet

permeability and the mass diffusion coefficient of the small molecular probes and described

structure effects on the leakage kinetics. Although improved, such leakage issues are not totally

resolved in fluorinated surfactants, for example, Skhiri et al. also highlighted the role of

fluorinated surfactant in the transport of small probes between neighboring droplets in fluorinert

oil (24). The authors found that the micelle concentration in the fluorinert oil phase impacts the

exchange kinetics in the emulsion.

Towards the development of a digital ELISA platform, we were able to optimize a system which

crucially decreases the fluorophore leakage from picoliter water-in-oil drops. With the addition

of sugar-derivatives in the aqueous phase, the signal is significantly conserved for the time

Page 5 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

6

required for enzymatic turnover of fluorogenic substrates. This method can address previously

mentioned fluorophore-related limitations and suggests new ways of achieving functional digital

microfluidic platforms. Moreover, we report the optimal conditions for digital sensing of

different fluorophores (resorufin, fluorescein and coumarin) where including sugar derivatives

decreases the fast fluorophore leakage from the droplets and maintains a significant signal level

over time. In addition, we demonstrate a microfluidic device that allows testing of the time-

dependent signal fidelity of various fluorophores in solution compartmentalized in mineral oil

with an appropriate surfactant.

Page 6 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

7

Experimental

Material

Amplex Ultra Red (AUR) substrate, horseradish peroxidase (HRP) and H2O2 were from the

ELISA Development Kit, Invitrogen (USA). FITC (fluorescein-based probe) was from Sigma-

Aldrich Inc. (USA). DiFMU (coumarin-based substrate) was purchased from Life Technologies

Corp. (USA). Mineral oil (paraffin heavy) and Tween80 were obtained from Thermo Fisher

Scientific Inc. (USA). Span80 (sorbitan monooleate) was obtained from Sigma-Aldrich Inc.

(USA). Sucrose, dextran (MW: 70 kDa), Fructose, Glucose, Galactose, Mannose, Arabinose,

Trehalose, Glycerol and bovine serum albumin (BSA) were purchased from Sigma Aldrich Inc.

(USA). Carboxymethyl-cellulose was from Lineco Inc. (USA). Polydimethylsiloxane (PDMS)

was purchased as Sylgard 184 from Dow Corning (USA). SU-8 was from Microchem, Newton,

MA (USA).

Chip fabrication

The microfluidic chip was fabricated by using standard photolithography and PDMS replica

molding protocols. SU-8 2050 resist was spun at 3,000 rpm on a 4-inch wafer and baked for 3

minutes at 65°C and 6 minutes at 95°C. The photoresist was exposed to UV light through a

chrome mask using a Karl Suss MA6 contact aligner (Suss, Germany) with an exposure dose of

approximately 190 mJ/cm2. The wafer was then cured at 65°C for 1 minute and 95°C for 6

minutes and developed for 5 minutes using SU-8 developer. Then the wafer was baked at 180°C

for 5 minutes. A 10:1 (base:linker) mixture of PDMS was added onto the mold and cured at

65°C for 3 hours. The final PDMS microchannel was bonded to a glass slide using an oxygen

plasma treatment and baked at 65°C for 12 hours as a final step.

Page 7 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

8

Droplet generation for emulsion stability characterization

A flow-focusing junction was designed and fabricated according to previous work (25). The

mineral oil was mixed with different concentration of surfactant combining Span80 and

Tween80 from 0 to 10% (w/w). The phases were injected at constant flow rate using glass

syringes (Hamilton Company, USA) and standard syringe pumps (Harvard apparatus, USA). The

syringes were connected to the device using peek tubing 20 (IDEX, Health & Science Corp.,

USA). 25ga (1/2”) luer stubs (Instech Laboratories Inc., USA) were used to connect the glass

syringe to the peek tube. The mineral oil and aqueous flow rates for generating droplets were

respectively 8μl/min and 2µl/min. The droplets were collected and imaged downstream in a large

chamber. The chamber thickness (50µm) was maintained by posts.

Compartmentalization array for signal fidelity assay

A two-dimensional array of parallel straight channels (30μm width) studded with lateral round-

shaped traps of 40μm width was designed. The aqueous phase with the fluorophore and the sugar

derivatives was injected using a standard syringe pump and was loaded in all lateral chambers by

dead-end filling. Then the aqueous phase injection was stopped and the mineral oil phase was

injected and maintained during the entire experiment using another syringe pump. Thus confined

droplets were produced in all lateral chambers. The two-dimensional array was time-lapse

imaged using a standard microscope (see below) and the fluorescence intensity in each droplet

was tracked following the frames in the sequence and quantified using ImageJ software. A

constant oil flow rate of 5µl/min was observed to result in quick fluorophore leakage and was

conserved for each experiment.

Page 8 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

9

In-tube emulsion

Water/Span80 phases (1:1 w/w) were vortexed in a centrifuge tube for 30-60 sec. The fractions

of remaining aqueous phase and opaque Span80 micelles were retrieved, measured and

normalized to their respective inputs.

Fluorescent signal generation

The stability of the water-in-mineral oil emulsion including surfactant was demonstrated using a

flow focusing junction (25, Fig. S1a) and the packed aqueous droplets of 40µm were imaged in a

large chamber (Fig. S1b). A concentration up to 10% of mixed surfactants in the mineral oil was

found to be sufficient to prevent the droplets to fuse over a long period of time. Amplex Ultra

Red (AUR) was chosen for its demonstrated high sensitivity as a substrate for enzyme

immunoassays (11, 14). It is a non-fluorescent substrate quickly transformed in the presence of

horseradish peroxidase (HRP) into a strong fluorescent resorufin derivative. The previously

mentioned rapid leakage through the droplets was highlighted using the fluorescent derived

product of 50µM AUR substrate transformed using HRP with 2µM H2O2. To enable optimal

HRP activity, Amplex Ultra Red was buffered at pH 6.0 which also conserves its maximum

fluorescence. FITC and DiFMU were resuspended in PBS, pH 7.4.

Imaging

Droplet generation, signal stability and signal leakage were characterized using 4x and 10x

objectives of a standard inverted fluorescence microscope (Eclipse Ti-U, Nikon Corp., JP). As

the droplets were produced at a high frequency, bright-field images were captured using a high-

speed camera at 5000 frames per second (Phantom V711, Vision Research Inc., USA) and

Page 9 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

10

processed on Phantom camera control software (Vision Research Inc., USA). The stabilized

emulsion was imaged in bright-field and fluorescence using a standard camera (DS-Qi1M, Nikon

Corp., JP) and processed on NIS-Element software (v3.22.01-2010, Nikon Corp., JP).

Page 10 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

11

Results

Using a microfluidic device integrating a flow-focusing junction prior to a large open chamber

(Fig. S1a-b), we first observed that rapid fluorophore leakage occurs in emulsions upon the

addition of stabilizing surfactant (a mix of Span80 and Tween80, 15) in the mineral oil phase

(Fig. S2a-b). The Amplex Ultra Red substrate (AUR), a widely used resorufin-based substrate

for horseradish peroxidase clearly shows this leakage behavior. We also observed that the

resorufin-derived fluorophores stayed in the oil phase and did not re-enter empty neighbouring

droplets. Reducing the concentration of surfactant decreased the fluorophore leakage as

previously described (24) but enabled the droplets to fuse.

To insure uniform and reproducible conditions, we characterized this fluorophore leakage from

aqueous drops using a microfluidic device that allowed generation of stable droplet volumes

independent of surfactant properties. The device consisted of a two-dimensional massively

parallel array of 40 µm wide semi-circular drop traps (Fig. 1a). By successively running through

the device the aqueous phase followed by the oil phase, monodisperse droplets were produced

and maintained in each trap of the array. This compartmentalization method was able to generate

more than 10’000 droplets per chip and also allowed the emulsion to be maintained independent

of the incorporation of surfactant. Moreover in this platform, droplets could be imaged over time

using epifluorescence microscopy, without significant motion and with control of the external

fluid convection and environment. Resorufin, initially encapsulated within the droplets was

observed to exit the droplets only in the presence of surfactant (Fig. 1b). Maintaining the oil flow

during the experiment convected away the exiting fluorophore at the droplet interface and

Page 11 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

12

increased the concentration gradient across the droplet, such that a higher flow rate led to an

increased droplet leakage rate (Fig. 1b-c).

Very interestingly, the addition of sucrose to the droplets mitigated the leakage of resorufin (Fig.

1b, d). After 450 seconds of imaging, signal fidelity was notably conserved up to 90% of the

initial signal intensity with the addition of sugars to the aqueous solution (Fig. 1b, bottom line;

Fig. 2a-b) compared to 25% in the absence of sugars (Fig. 1b, middle line; Fig. 2a-b). Note that

the fluorophore leakage or addition of sucrose did not induce a change of the droplet size over

time. Increasing the fluorophore solubility in the aqueous phase using dimethyl-sulfoxide

(DMSO) or saturating the oil using non-fluorescent AUR substrate which both were expected to

slow down the leakage rate did not prevent the fluorophore leakage (data not shown).

In order to further characterize the impact of sugar on the fluorophore leakage, different

concentrations and sugar-derivatives were tested. We found that inhibition of fluorophore

leakage positively correlated with the amount of sucrose included in the aqueous phase (Fig. 2a)

and that different sugar derivatives (sucrose, dextran but not cellulose) displayed similar

stabilization characteristics (Fig. 2a-b). Although dextran was tested at a higher dilution (5%

w/v) than sucrose (25% w/v), the fluorescent signal stability of the droplets was comparable in

both cases. Finally, the addition of carboxymethyl-cellulose resulting in a more viscous phase

did not prevent the fluorophore leakage and these conditions resulted in the same final leakage

rate as in the standard condition without any sugar additive (Fig. 2c).

Page 12 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

13

To assess which components of the surfactant were responsible for inducing the resorufin

fluorophore leakage, we repeated the experiments and separately added the two surfactant

fractions (Span80 and Tween80) to the mineral oil. It appeared that the Span80 fraction played a

major role in accelerating leakage (Fig. 2c) even if the combination of both fractions resulted in a

much higher leakage rate.

We also aimed to understand if sugar derivatives preferentially have an effect on either the

ability to generate surfactant micelles that would enhance leakage or on the partitioning of the

fluorophore in the aqueous phase. To address this question, we decided to generate emulsions

using water and Span80 only (without fluorophore or mineral oil). Without a fluorophore acting

as the signal reporter, the microfluidics platform could not be used for imaging and a larger

quantity of emulsion that was more easily visualized was generated by vortexing the two phases

in a centrifuge tube. The effect of an additive in the aqueous phase was reported by quantifying

the remaining aqueous phase and the dense emulsion volume fraction generated by vortexing the

aqueous and Span80 phases (Fig. S3). Note that this approach directly probes surfactant micelle

stability within the aqueous phase and separately from the stability of an aqueous emulsion in oil.

We observed different emulsion fractions using Span80 alone or Span80 combined with the

sugar derivatives in the aqueous phase, as well as saturated aqueous solutions of bovine serum

albumin (BSA) or 5% carboxymethyl-cellulose. Results confirmed a dispersive effect of the

sugar-derivatives on the emulsion as the dense emulsion volume fraction was significantly

diminished by the sugar additives (Fig. 2d and Fig. S4). Note that the Span80/water emulsion

was not generated due to a stabilized air/water interface and could be produced under vacuum

and suspended in mineral oil without observing any structural changes under the microscope.

Page 13 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

14

Finally, we found that the sugar additive impact on the fluorophore leakage was specific to the

resorufin fluorophore and different optimal additives and conditions were found to minimize the

leakage of other small fluorophores from droplets. Fluorescein isothiocyanate (FITC) and

difluoro-hydroxy-methylcoumarin (DiFMU) were tested using the microfluidic platform as

previously described. For FITC, the addition of surfactant alone already inhibited fluorophore

leakage compared to the condition in which both sugar and surfactant were absent (Fig. 3a). Yet

the highest signal fidelity was obtained by combining the sugar additives with no surfactant.

DiFMU, showed the highest maintenance within droplets in the absence of surfactant and sugar-

derivatives (Fig. 3b). When surfactant was present, the addition of sucrose did not prevent

DiFMU leakage while the addition of dextran improved signal fidelity. A summary of the effects

on the signal stability of the different fluorophores using each additive in the aqueous phase

reported as the maximum leakage rate is provided in the supplementary information (Table S1).

Page 14 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

15

Discussion

Several previous studies have highlighted the fundamental limitation imposed by the diffusion of

small fluorophores through the interface of water-in-oil emulsions (19-22). Some methods

employed to reduce fluorophore leakage include generating double emulsions (19) or separating

droplets by gas bubbles in a straight channel. Chemical modification of specific fluorophores has

also been introduced to increase stability (20) and new alternatives to synthesized surfactant are

now required to develop cost and time effective drop-based devices (26). Although effective,

these solutions greatly increased both the reagent cost and the platform complexity.

In this work, we designed a leakage quantification experiment presenting well-controlled

conditions to distinguish the role of the different emulsion components. We confirmed that the

presence of surfactant increases the leakage of certain small fluorophores such as resorufin

derivatives across droplets boundaries. The addition of a constant flow passing between the

droplets also significantly increased the leakage. Renewing continuously the mineral oil and

surfactant phase using constant injection enabled us to amplify the fluorophore leakage rate and

to perform time-lapse experiments with a limited number of secondary effects. Using such a

platform enabled testing various additives for inhibition of fluorophore leakage induced by the

surfactant.

Our results suggest that the addition of short-chained sugar derivatives inhibit the fluorophore

leakage and thus help to conserve the signal over time for digital quantification through enzyme

immunoassays. The sugar additives may act in two main ways to reduce leakage: (i) by directly

interacting with and stabilizing the fluorophore within the aqueous phase (i.e. enhancing the

Page 15 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

16

equilibrium partition coefficient), or (ii) interacting with the surfactant to either (a) prevent

formation of micelles containing fluorophore, or (b) reduce the rate of micelle shedding overall.

Since the sugar additives were able to reduce the stability of water/Span80 emulsions (i.e. lower

emulsion fraction of our in-tube experiments), they likely directly interact with the Span80

surfactant, supporting mechanism (ii) described above. As Span80 is known to generate reverse

micelles (27-29) at the droplet surface, it is possible that the sugar derivatives directly disrupt the

surfactant organization at the droplet interface reducing the formation and shedding of micelles

which contain fluorophores. Note that these potential surfactant micelles were not directly

observed and therefore would be sub-micron sized. Moreover we did not observe the formation

of a surfactant bilayer at the droplet interface.

On the other hand, kinetics of leakage has also been related to direct partitioning between the

two phases (23, 24). For example, in the model of Chen et al., reducing the fluorophore

permeability was predicted to best inhibit leakage; while reducing the diffusivity using a viscous

aqueous solution like dissolved carboxymethyl-cellulose had a less prominent effect. Using

carboxymethyl-cellulose was also demonstrated to be inefficient for inhibiting the leakage of

fluorescein-derivatives in another study (19). Our results indicate that future leakage models

should consider additional effects, in which the surfactant itself acts to facilitate fluorophore

permeation across the water-oil interface, which may especially play a role when non-

equilibrium conditions are present. In our experiments, droplets are physically separated in the

microfluidic array and the oil flow convects away exiting fluorophores / micelles and thus

modifies their partition since the system never tends to equilibrium. As a consequence, when

Page 16 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

17

linking the exposed droplet intensity decays to current published models, care should be taken to

understand the previously described boundary conditions that govern the kinetics of leakage.

The demonstrated interaction between sugars and the Span80 fraction highlights the new field of

engineering interfering molecules to solubilize small probes in the aqueous phase, reduce their

permeability and prevent transport from a drop. High concentrations of monosaccharides or

disaccharides enabled reduced generation of water/Span80 emulsions with a similar efficiency.

Thus the choice of the best non-branched or short branched sugar additive for signal

conservation might only depend on its solubility in water. Therefore sucrose or fructose that have

the highest solubility might be best suited. In addition dextran, a polysaccharide, exhibited a

particular high performance since it still reduced the resorufin leakage and the Span80 micelles

at lower concentration than other sugar derivatives. Sugar alcohols showed a lower potential at

equal concentration than sugars to reduce the Span80 emulsion. This shows that interfering with

the surfactant is related to the precise chemical structure of saccharides. The similar dispersive

effect of BSA observed on Span80 (but still smaller than sugar derivatives) might highlight

potential causes of its previously reported benefits in emulsions (19, 24).

Differences in leakage kinetics for fluorescein and coumarin derivatives show that each emulsion

system has unique characteristics depending on the chemical structure and the charge of the

fluorophore and how they specifically interact with surfactants. Ultimately, multiplexing a digital

assay using resorufin, fluorescein and coumarin-based fluorophores might require a compromise

in the addition of surfactant and sugar additives. Based on our results, a combination of these

Page 17 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

18

dyes would optimally combine surfactant and 5% dextran to best conserve the signal fidelity of

all three fluorophores.

Page 18 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

19

Conclusion

We report a new technique to overcome the poor signal stability of amphipathic readout

fluorophores in water-in-oil picoliter reaction confinement systems. The addition of sugar

derivatives in the aqueous phase enables conservation of the resorufin-based signal stability over

a time period necessary for enzymatic accumulation of fluorophore and imaging of droplets.

Improved signal stability using available cost-effective reagents unlocks new opportunities for

digital water-in-oil droplet immunoassays. Moreover this approach can be expanded to any use

of resorufin-based fluorophores in a multi-phase microfluidic format.

Page 19 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

20

Figure 1. Characterization of the microfluidic platform for quantifying fluorophore leakage in an

emulsion (a) Picture of compartmentalized aqueous droplets in the two-dimensional parallel

array of lateral traps (b) Characterization of fluorophore leakage under three conditions: no

surfactant/no sugar, surfactant/no sugar and surfactant/25% sucrose. (c) Fluorophore leakage

quantification at different oil flow rates (d) Schematic of resorufin leakage prevention from

droplets induced by sucrose in the presence of surfactant.

Page 20 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

21

Figure 2. Quantification of fluorophore leakage reduction by different sugar derivatives. (a) The

aqueous drops were loaded with the fluorophore along with different concentrations of sucrose.

(b) Same as above using dextran instead of sucrose. (c) Same as above using carboxymethyl-

cellulose additive. (d) In-tube water/Span80 emulsion (1:1 w/w).

Page 21 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

22

Figure 3. Leakage characterization of two other small fluorescent probes: (a) Fluorescein

(50µM) (b) DiFMU (50µM).

Page 22 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

23

AUTHOR INFORMATION

Corresponding Author

Dino Di Carlo: dicarlo@seas.ucla.edu; Department of Bioengineering, University of California,

Los Angeles, 410 Westwood Plaza, Los Angeles 90095, CA, USA.

Present Addresses

†If an author’s address is different than the one given in the affiliation line, this information may

be included here.

†Patrick A. Sandoz: Global Health Institute, School of Life Sciences, EPFL, VDG, Station 19,

Lausanne 1015, Switzerland.

††Aram J Chung: Mechanical, Aerospace and Nuclear Engineering, Rensselaer Polytechnic

Institute, Troy, NY 12180

Page 23 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

24

Funding Sources

This work was supported in part by National Science Foundation grant #1332275.

ACKNOWLEDGMENT

We thank P. Renaud and A. Ozcan for helpful discussions and guidance, G. van der Goot for

providing some reagents and S. Friebe for critical proofreading of the manuscript.

SUPPLEMENTARY INFORMATION AVAILABLE

Experimental details for water-in-oil droplets generation and emulsion stabilization, resorufin

leakage in the presence of surfactant, a comparison of other various sugar derivatives on in-tube

emulsion and a summary of the leakage characterization for resorufin, fluorescein and coumarin

derivatives. This information is available free of charge via the Internet at http://pubs.acs.org/.

ABBREVIATIONS

ELISA, enzyme-linked immunosorbent assay, HRP, horseradish peroxidase, AUR, Amplex Ultra

Red, FITC, Fluorescein isothiocyanate, DiFMU, 6,8-Difluoro-7-Hydroxy-4-Methylcoumarin,

PDMS, polydimethylsiloxane, PFOA, perfluorooctanoic acid, PFOS, perfluorooctane

sulphonate.

Page 24 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

25

REFERENCES

(1) Teh, S. Y.; Lin, R.; Hung, L. H. and Lee, A. P. Droplet microfluidics. Lab Chip 2008, 8,

198–220

(2) Guo M. T.; Rotem, A.; Heymana, J. A. and Weitz, D. A. Droplet microfluidics for high-

throughput biological assays. Lab Chip 2012, 12, 2146–2155

(3) Niu, X. and deMello, A. J. Building droplet-based microfluidic systems for biological

analysis. Biochem. Soc. Trans. 2012, 40, 615–623

(4) Vogelstein, B. and Kinzler, K. W. Digital PCR. Proc. Natl. Acad. Sci. U. S. A. 1999, 96,

9236–9241

(5) Hatch, A. C.; Fisher, J. S.; Tovar, A. R.; Hsieh, A. T.; Lin, R.; Pentoney, S. L.; Yang D. L.

and Lee, A. P. 1-Million droplet array with wide-field fluorescence imaging for digital PCR. Lab

Chip 2011, 21, 3838-3845

(6) Gansen, A.; Herrick, A. M.; Dimov, I. K.; Lee, L. P. and Chiu, D. T. Digital LAMP in a

sample self-digitization (SD) chip. Lab Chip 2012, 12, 2247–2254

(7) Shen, F.; Du, W.; Kreutz, J. E.; Fok, A. and Ismagilov, R. F. Digital PCR on a SlipChip. Lab

Chip 2010, 10, 2666–2672

(8) Shen, F.; Sun, B.; Kreutz, J. E.; Davydova, E. K.; Du, W.; Reddy, P. L.; Joseph, L. J. and

Ismagilov, R. F. Multiplexed quantification of nucleic acids with large dynamic range using

multivolume digital RT-PCR on a rotational SlipChip tested with HIV and hepatitis C viral load.

J. Am. Chem. Soc. 2011, 133, 17705–17712

(9) Hindson, B. J.; Ness, K. D.; Masquelier, D.; Belgrader, P.; Heredia, N. J.; Makarewicz, A. J.;

Bright, I. J.; Lucero, M. Y.; Hiddessen, A. L.; Legler, T. C.; et al. High-throughput droplet

digital PCR system for absolute quantitation of DNA copy number. Anal. Chem. 2011 83, 8604–

8610

(10) Zhu, Z.; Zhang, W.; Leng, X.; Zhang, M.; Guan, Z.; Lu J. and Yang, C. J. Highly sensitive

and quantitative detection of rare pathogens through agarose droplet microfluidic emulsion PCR

at the single-cell level. Lab Chip 2012, 12, 3907–3913

(11) Agresti, J. J.; Antipov, E.; Abate, A. R.; Ahn, K.; Rowat, A. C.; Baret, J. C.; Marquez, M.;

Klibanov, A. M.; Griffiths, A. D. and Weitz, D. A. Ultrahigh-throughput screening in drop-based

microfluidics for directed evolution. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 4004–4009

(12) Bai, Y.; He, X.; Liu, D.; Patil, S. N.; Bratton, D.; Huebner, A.; Hollfelder, F.; Abell, C. and

Huck, W. T. S. A double droplet trap system for studying mass transport across a droplet-droplet

interface. Lab Chip 2010, 10, 1281–1285

Page 25 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

26

(13) Pekin, D.; Skhiri, Y.; Baret, J. C.; Le Corre, D.; Mazutis, L.; Ben Salem, C.; Millot, F.; El

Harrak, A.; Hutchison, J. B.; Larson, J. W.; et al. Quantitative and sensitive detection of rare

mutations using droplet-based microfluidics. Lab Chip 2011, 11, 2156–2166

(14) Rissin, D. M.; Kan, C. W.; Campbell, T. G.; Howes, S. C.; Fournier, D. R.; Song, L.; Piech,

T.; Patel, P. P.; Chang, L.; Rivnak, A. J.; et al. Single-molecule enzyme-linked immunosorbent

assay detects serum proteins at subfemtomolar concentrations. Nat. Biotechnol. 2010, 28, 595–

599

(15) Baret, J. C.; Surfactants in droplet-based microfluidics. Lab Chip 2012, 12, 422–433

(16) Mazutis, L.; Araghi, A. F.; Miller, O. J.; Baret, J. C.; Frenz, L.; Janoshazi, A.; Taly, V.;

Miller, B. J.; Hutchison, J. B.; Link, D.; et al. Droplet-Based Microfluidic Systems for High-

Throughput Single DNA Molecule Isothermal Amplification and Analysis. Anal. Chem. 2009,

81, 4813–4821

(17) Holtze, C.; Rowat, C.; Agresti, J. J.; Hutchison, J. B.; Angilè, F. E.; Schmitz, C. H. J.;

Köster, S.; Duan, H.; Humphry, K. J.; Scanga, R. A.; et al. Biocompatible surfactants for water-

in-fluorocarbon emulsions. Lab Chip 2008, 8, 1632–1639

(18) Kostov, G.; Boschet, F.; Ameduri, B. Original fluorinated surfactants potentially non-

bioaccumulable. J. Fluorine Chem. 2009, 130, 1192-1199

(19) Courtois, F.; Olguin, L. F.; Whyte, G.; Theberge, A. B.; Huck, W. T. S.; Hollfelder F. and

Abell, C. Controlling the Retention of Small Molecules in Emulsion Microdroplets for Use in

Cell-Base Assays. Anal. Chem. 2009, 81, 3008–3016

(20) Woronoff, G.; El Harrak, A.; Mayot, E.; Schicke, O.; Miller, O. J.; Soumillion, P.; Griffiths,

A. D. and Ryckelynck, M. New Generation of Amino Coumarin Methyl Sulfonate-Based

Fluorogenic Substrates for Amidase Assays in Droplet-Based Microfluidic Applications. Anal.

Chem. 2011, 83, 2852–2857

(21) Marcoux, P. R.; Dupoy, M.; Mathey, R.; Novelli-Rousseau, A.; Heran, V.; Morales, S.;

Rivera, F.; Joly, P. L.; Moy, J. P.; Mallard, F. Micro-confinement of bacteria into w/o emulsion

droplets for rapid detection and enumeration. Colloids Surf., A 2011, 377, 54–62

(22) Wu, N.; Courtois, F.; Zhu, Y.; Oakeshott, J.; Easton, C.; Abell, C. Management of the

diffusion of 4-methylumbelliferone across phases in microdroplet-based systems for in vitro

protein evolution. Electrophoresis 2010, 31, 3121–3128

(23) Chen, Y.; Gani, A. W.; Tang, S. K. Y. Characterization of sensitivity and specificity in

leaky droplet-based assays. Lab Chip 2012, 12, 5093-5103

(24) Skhiri, Y.; Gruner, P.; Semin, B.; Brosseau, Q.; Pekin, D.; Mazutis, L.; Goust, V.;

Kleinschmidt, F.; El Harrak, A.; Hutchison, J. B.; et al. Dynamics of molecular transport by

surfactants in emulsions. Soft Matter 2012, 8, 10618–10627

Page 26 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

27

(25) Edd, F. F.; Di Carlo, D.; Humphry, K. J.; Köster, S.; Irimia, D.; Weitz, D. A. and Toner, M.

Controlled encapsulation of single-cells into monodisperse picolitre drops. Lab Chip 2008, 8,

1262–1264

(26) Holt, D. J.; Payne, R. J.; Abell, C. Synthesis of novel fluorous surfactants for microdroplet

stabilisation in fluorous oil streams. J. Fluorine Chem. 2010, 131, 398–407

(27) Sela, Y.; Magdassi, S.; Garti, N. Release of markers from the inner water phase of W/O/W

emulsions stabilized by silicone based polymeric surfactants, J. Controlled Release 1995, 33, 1-

12

(28) Cheng, J.; Chen, J. F.; Zhao, M.; Luo, Q.; Wen, L. X.; Papadopoulos, K. D. Transport of

ions through the oil phase of W1/O/W2 double emulsions, J. Colloid Interface Sci. 2007, 305,

175–182

(29) Kato, K.; Walde, P.; Koine, N.; Ichikawa, S.; Ishikawa, T.; Nagahama, R.; Ishihara, T.;

Tsujii, T.; Shudou, M.; Omokawa Y. and Kuroiwa, T. Temperature-Sensitive Nonionic Vesicles

Prepared from Span80 (Sorbitan Monooleate). Langmuir 2008, 24, 10762-10770

Page 27 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960

28

Sugar Additives Improve Signal Fidelity for Implementing Two-phase

Resorufin-based Enzyme Immunoassays

Patrick A. Sandoz†, Aram J. Chung††, Westbrook M. Weaver, Dino Di Carlo

*

Department of Bioengineering, University of California, Los Angeles, 90095, CA, USA.

GRAPHIC TABLE OF CONTENT:

Page 28 of 28

ACS Paragon Plus Environment

Langmuir

123456789101112131415161718192021222324252627282930313233343536373839404142434445464748495051525354555657585960