Fluorinated Pickering Emulsions with Nonadsorbing Interfaces for

Page 1 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Analytical Chemistry

Fluorinated

Pickering

Emulsions

with

Non-

adsorbing Interfaces for Droplet-based Enzymatic Assays Ming Pan1, Fengjiao Lyu2, and Sindy K. Y. Tang*,2 1

Department of Material Science and Engineering, Stanford University, Stanford, California

94305-2004, United States 2

Department of Mechanical Engineering, Stanford University, Stanford, California 94305-2004,

United States * Email: [email protected] * Fax: 650-723-1748

ACS Paragon Plus Environment

1


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 24

ABSTRACT

This work describes the use of fluorinated Pickering emulsions with non-adsorbing interfaces in droplet-based enzymatic assays. State-of-the-art droplet assays have relied on one type of surfactants consisting of perfluorinated polyether and polyethylene glycol (PFPE-PEG). These surfactants are known to have limitations including the tedious synthesis and inter-drop molecular transport which leads to the cross-contamination of droplet contents. Previously we have shown that replacing surfactants with nanoparticles as droplet stabilizers mitigate inter-drop transport of small molecules. The non-specific adsorption of enzymes on nanoparticle surface, however, could cause structural changes in enzymes and consequently the loss of enzymatic activity. To overcome such challenge, we render nanoparticle surface non-adsorbing to enzymes by in-situ adsorption of polyethylene glycol (PEG) on particle surfaces. We show that enzyme activities are preserved in droplets stabilized by PEG-adsorbed nanoparticles, and are comparable with those in drops stabilized by PFPE-PEG surfactants. In addition, our nonadsorbing Pickering emulsions successfully prevent inter-drop molecular transport, thereby maintaining the accuracy of droplet assays. The particles are also simple and economical to synthesize. The PEG adsorbed nanoparticles described in this work are thus a competitive alternative to the current surfactant system, and can potentially enable new droplet-based biochemical assays.

KEYWORDS: Pickering emulsion; droplet microfluidics; nanoparticles; amphiphilicity; biocompatibility


2

Page 3 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


INTRODUCTION This paper describes the generation of amphiphilic silica nanoparticles with non-adsorbing interfaces for the stabilization of aqueous drops in fluorinated oils for use in enzymatic assays in droplet microfluidics. Droplet microfluidics, in which droplets act as individual reactors, has enabled a wide range of applications in high-throughput screening and diagnostics.1,2,3 One key application area is the directed evolution of enzymes.2,4,5 Such assays typically involve the compartmentalization of a library of mutant enzymes in droplets along with a fluorogenic substrate. The enzymatic activity of different mutants is measured optically from the drops. Drops exhibiting high activity are sorted for another round of mutagenesis and screening. The success of these assays hinges on the following two criteria: 1) the encapsulation of enzymes in droplets does not degrade enzyme activity, and 2) the enzyme activity can be quantified accurately in droplets. State-of-the-art droplet-based assays have predominately relied on a single type of surfactants (PFPE-PEG, or “EA-surfactants”), which is a triblock copolymer consisting of fluorinated alkyl domains as the fluorophilic tail and polyethylene glycol (PEG) domains as the hydrophilic head.6 Fluorophilic tails are necessary as fluorinated solvents are often used as the continuous phase due to their gas permeability and biocompatibility.6 The presence of the PEG domain is critical for preventing non-specific adsorption of proteins and enzymes to the droplet interface and to maintain their activities.6-9 It is also important for preventing adsorption of other biomolecules such as DNA and RNA, and for the biocompatibility with the growth of suspending cells in drops.6 Nonetheless, the synthesis and purification of such fluorinated surfactants are tedious and costly. Several new approaches have been developed to reduce the synthetic complexity. For example, the fluorinated and PEG domains were assembled in-situ at


3


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 4 of 24

the water-oil interface via electrostatic interactions.10 A serious limitation of these surfactants, however, is that they have been shown to cause cross-contaminations of small, hydrophobic molecules among droplets.11-15 Such surfactant-assisted inter-drop molecular transport (referred to as “leakage” herein) presents a key obstacle to the accurate interrogation of enzymatic reactions, which often employ small, hydrophobic molecules as the fluorophore in fluorogenic substrates. In our previous study,16 we showed that the leakage of model fluorophore resorufin was mitigated by replacing surfactants with partially fluorinated nanoparticles (“F-SiO2 NPs”) as droplet stabilizers to form Pickering emulsions. Since the nanoparticles were irreversibly adsorbed at the interface and they did not form reverse micelles, a major pathway for leakage was eliminated. We further showed that these F-SiO2 NPs-stabilized drops exhibited biocompatibility with bacteria, and provided a rigid interface for the adhesion and growth of anchorage-dependent mammalian cells. Nevertheless, it is known that enzymes and proteins adsorb non-specifically to silica surfaces and can become denatured.17-21 Replacing PFPE-PEG surfactants with silica nanoparticles could thus lead to diminished enzyme activities and result in ineffective droplet-based enzymatic assays. In order to maintain enzymatic activity and assay accuracy, the objective of this paper is to develop an approach that prevents 1) non-specific adsorption and subsequent degradation of enzymes, and 2) the leakage of fluorophores used in fluorogenic substrates. We modify our Pickering emulsion system by introducing PEG into the dispersed phase. The F-SiO2 NPs are pre-dispersed in the continuous phase. As drops are generated, the F-SiO2 NPs adsorb to the water-oil interface and the PEG adsorbs onto the surface of the F-SiO2 NPs from within the drops. We refer to these particles as “PEGads-F-SiO2 NPs” in the rest of the paper.


4

Page 5 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


EXPERIMENTAL SECTION Materials. All chemicals were used as purchased without purification. Absolute ethanol (99%), tetraethyl orthosilicate (TEOS) (98%), ammonium hydroxide solution (28%), poly-ethylene glycol (PEG, MW=8000), fluorescein, sodium salts of resazurin and resorufin were purchased from Sigma-Aldrich. 1H, 1H, 2H, 2H-perfluorooctyltriethoxysilane (FAS) (97%) was purchased from Fisher Scientific. Methoxyl PEG fluorescein (mPEG-Fluorescein, MW=5000) was purchased from Nanocs Inc. Methoxyl PEG silane (MW=1000) was purchased from Laysan Bio. Inc.. Fluorescein diphosphate (FDP), bacterial alkaline phosphatase (BAP), and Amplex Red enzyme kit were purchased from Life Technologies. Synthesis of 100 nm F-SiO2 NPs. 100 nm F-SiO2 NPs were synthesized according to the method published previously with modifications.16 3.57 mL of TEOS was added to a solution mixture containing 50 mL of ethanol (EtOH), 1 mL of deionized water, and 1.43 mL of NH4OH (28 wt %). The solution was stirred vigorously at room temperature for 12 hours. 100 µL of FAS was then added directly to every 3 mL of the synthesized SiO2 NPs solution obtained above, followed by vigorous stirring at room temperature for 60 min. EtOH was added to dilute the reacting solution to terminate the reaction, and the particles were washed by centrifugation at 10,000 rpm for 20 min. After three cycles of washing, the supernatant was removed and the resulting particles were desiccated overnight. Functionalization of F-SiO2 NPs with PEGylated silane. 300 µL of FAS and either 80 µL or 750 µL of mPEG-silane in EtOH (10 mg/mL) were mixed and then added to 3 mL of synthesized SiO2 NPs. The resulting mixture was stirred at room temperature for 30 min and the PEGylated NPs were isolated following the same procedure as that described for F-SiO2 NPs.


5


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 6 of 24

Droplet generation. Monodisperse microdroplets were generated from flow-focusing devices. The continuous phase contained either 1.5% (w/w) 100-nm F-SiO2 NPs dispersed in HFE-7500 or 2% (w/w) EA-surfactant in HFE-7500. For the dispersed phase, the composition is listed in Table S1 (supporting information). The flow rates of the continuous phase and the two streams of dispersed phase were fixed at 1.0 mL/hr, 0.1 mL/hr and 0.1 mL/hr respectively. Leakage test for Amplex Red assay with different [HRP]. Positive and negative droplets were generated separately. Continuous phase contained 1.5 % (w/w) 100 nm F-SiO2 NPs dispersed in HFE-7500. Positive droplets contained 75 µM Amplex Red reagent, 5 mU/mL HRP, 4 mg/mL PEG, 1 mM H2O2 and 10 µM fluorescein. Negative droplets contained 75 µM Amplex Red reagent, 0.1 mU/mL HRP, 4 mg/mL PEG, 1 mM H2O2 and 2 µM fluorescein. The droplets were collected in two Eppendorf tubes. Excess NPs in continuous phase was removed by washing with FC-40 three times. FC-40 was used instead of HFE-7500 to minimize partitioning of resorufin into the continuous phase, which was observed for fluorinated solvents that contained aliphatic ether groups.16 The positive droplets and negative droplets were then mixed at 1:1 ratio. The droplet mixture was incubated at room temperature (293 K) for 4 hours. In a separate control experiment, positive and negative droplets stabilized by EA-surfactant were mixed. The procedures were identical to that for NPs stabilized droplets except the continuous phase contained 2% EA-surfactant in HFE-7500, and no washing step was involved before the droplets were mixed.


6

Page 7 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


RESULTS AND DISCUSSIONS Droplet stability. We generated drops using a flow-focusing nozzle with one inlet for the continuous phase and two inlets for the dispersed phase (Fig. 1).22 The continuous phase contained 1.5% (w/w) of 100nm F-SiO2 NPs dispersed in fluorinated solvent HFE-7500. For the dispersed phase, one inlet was used for introducing an aqueous stream containing enzyme and PEG, and the other inlet was used for introducing an aqueous stream containing a fluorogenic substrate and an internal standard for subsequent fluorescence quantification. The two streams came into contact ~ 850 µm upstream of the flow-focusing nozzle. At the flow rates used, the contact time was approximately 18 milliseconds, ~ 105 times shorter than that needed for the completion of the enzymatic reactions used in this work (~ 1 hour). The contents of these two streams were thoroughly mixed after being encapsulated into droplets only (Fig. S1). This on-chip mixing ensured that the enzymatic reactions took place primarily within the droplet. The final concentration of PEG in the drops was fixed at 0.4% (w/v). Fig. 1b shows that the drops formed were monodisperse and stable against coalescence. The presence of enzyme and PEG in the disperse phase did not destabilize the drops. They were stable against coalescence under typical droplet manipulation conditions, such as washing and dilution with another fluorinated solvent FC-40, reinjection into narrow channels, and heating at 90 °C for 10 minutes (Fig. S2). We had anticipated these droplets to be highly stable because the nanoparticles and PEG were pre-dispersed in different phases before droplet generation, and they interacted after the formation of drops only. The presence of PEG in the drops did not affect the amphiphilicity of the particles, as PEG could not interact with the portion of the particles


7


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 8 of 24

exposed to the continuous phase. The concentration of PEG in the aqueous phase also did not affect the irreversible adsorption of F-SiO2 NPs to the water-oil interface.

PEG adsorption onto particle surface. To probe the adsorption of PEG onto particle surface, we measured the spatial distribution of PEG within the droplets stabilized by F-SiO2 NPs and EA-surfactants respectively (Fig. 2). For this figure only, we used fluorescein-tagged PEG (mPEG-fluorescein, MW=5000) in the dispersed phase. Since fluorescein was covalently linked to PEG, we can estimate the concentration of PEG at a given region by measuring the fluorescence intensity distribution. When the droplets were stabilized by EA-surfactant, the fluorescence intensity was maximum in the center of the drops due to their spherical shape (Fig. 2c), suggesting that PEG molecules were uniformly distributed inside the drop. This distribution was expected as the head group of the EA-surfactant already contained PEG, which prevented further adsorption and concentration of PEG at droplet interface. When the drops were stabilized by F-SiO2 NPs, the fluorescence intensity was maximum at the droplet interface forming fluorescent rims. A line scan across the image of a drop shows two intensity peaks which were separated by a distance equal to the droplet diameter (~65 µm, Fig. 2c). Such intensity profile indicates that PEG molecules were concentrated at the droplet interface. It was not due to the increased scattering of light by the particles as a drop containing fluorescein only did not produce such fluorescent rims (Fig. S3c). The presence of enzymes did not affect the rim formation in nanoparticle-stabilized drops, even at high enzyme concentrations (~33.75 U/mL) (Fig. S3b). It indicates that the adsorbed PEG could not be displaced by enzyme molecules, as the PEG would be distributed uniformly within the droplet otherwise.


8

Page 9 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


This result is consistent with our expectation: since the molecular weight of the PEG (MW~5000-8000) was smaller than that of the enzymes used (MW~89000 for bacterial alkaline phosphatase23 and MW~44000 for horseradish peroxidase24), the diffusion and adsorption of PEG onto nanoparticle surfaces should be faster than that of the enzymes. In addition, the adsorption of PEG is known to be driven by the formation of hydrogen bonds in the presence of polar groups such as silanol groups on surfaces.25-27 Formation of hydrogen bond between PEG chains and silanol groups is expected in our case since our silica nanoparticles were only partially fluorinated (see XPS characterization in our previous paper16). Subsequent enzyme adsorption on particle surface is prevented due to steric repulsions between the PEG chains and enzymes. Such repulsions are known to predominate over attractive interactions between enzymes and nanoparticle surface.7,28

Restoration of enzymatic activities. To test if the presence of PEG would preserve enzymatic activity, we investigated two representative assays involving enzymes alkaline phosphatase and horseradish peroxidase respectively. In the first assay, we used Bacterial Alkaline Phosphatase (BAP) to catalyze the hydrolysis of fluorescein diphosphate (FDP) to produce a green fluorescent product fluorescein. We introduced resorufin (0.01 mg/mL, 42.5 µM) to each drop as an internal standard to normalize the fluorescence intensity of fluorescein (see Table S1 for detailed compositions of the assay). We normalized fluorescence intensity by calculating the fluorescence intensity ratio between fluorescein and resorufin. Due to the non-overlapping excitation/emission wavelengths between fluorescein (494 nm/521 nm) and resorufin (571 nm/585 nm), the presence of resorufin did not interfere with the fluorescence of fluorescein.


9


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 10 of 24

Fig. 3a shows the time evolution of fluorescence signal for fluorescein from F-SiO2 NPsstabilized droplets with or without PEG. In the absence of PEG, the fluorescence intensity of fluorescein did not increase over time, indicating the possible deactivation of BAP. The addition of PEG into F-SiO2 NPs-stabilized drops greatly enhanced the fluorescence turn-on rate, which was comparable with the rates in drops stabilized by EA-surfactant and in bulk assays performed in a 96-well-plate (Fig. S4). The addition of PEG did not have significant effect on the reaction rate in both surfactant-stabilized drops and in bulk samples (Fig. S4). Similar results were observed for horseradish peroxidase (HRP) where Amplex Red reagent ((10-acetyl-3,7dihydroxyphenoxazine) was oxidized to resorufin in the presence of hydrogen peroxide and HRP (Fig. 3b). 10 µM fluorescein was used as an internal standard here. Although HRP was generally believed to be an efficient enzyme, its activity in F-SiO2 NPs-stabilized droplets was lost in the absence of PEG. The addition of PEG in F-SiO2 NPs-stabilized droplets was effective in restoring enzyme activity to a level comparable with state-of-the-art EA-surfactant systems. Our results are consistent with previous reports that coating solid surfaces with PEG prevents undesirable non-specific adsorption of proteins and subsequent degradation of their structures and/or functions.9,29-33

Advantages of chemisorption of PEG over covalent PEG grafting on particle surface. To coat the nanoparticles with PEG, there are two plausible approaches: 1) covalent grafting of PEG onto nanoparticles prior to droplet generation, followed by the dispersion of these particles into the continuous phase. The dispersion of particles in the continuous phase is preferable to the dispersion in the dispersed phase to avoid undesirable interactions between particles and droplet contents. 2) Chemisorption of free PEG molecules from the dispersed phase onto F-SiO2 NPs


10

Page 11 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


originally dispersed in the continuous phase after droplet formation. The first approach was not feasible in our system because high PEG density critical for preserving enzyme function also made the particles too hydrophilic to be dispersible in the fluorinated oil. We had grafted PEG chains on the surface of F-SiO2 NPs with different coating densities by the hydrolysis of PEGlinked silane (mPEG-silane). The PEG density at the interface was adjusted by varying the concentration of precursor mPEG-silane in the synthesis of the F-SiO2 NPs (see experimental section). We used these PEGylated particles (referred to as “PEGcovalent-F-SiO2 NPs” thereafter) as droplet stabilizer to test BAP activity inside the drops (Fig. S5). Fluorescence measurement indicated that BAP activity was preserved only when we used high concentrations of precursor mPEG-silane, which corresponded to a high PEG graft density on particle surfaces (Fig. S5 (iii)). Unfortunately, these highly PEGylated PEGcovalent-F-SiO2 NPs also aggregated in HFE-7500, as the PEGylation rendered the particles hydrophilic and not dispersible in fluorinated solvents. This aggregation prevented the use of such particles in droplet microfluidics as they caused clogging of the channel. Lowering PEG grafting density made these PEGcovalent-F-SiO2 NPs dispersible in HFE-7500. Such particles could not maintain BAP activity, however (Fig. S5 (ii)). Here, the trade-off between the dispersibility of particles and their ability to preserve enzyme activity arose from the fact that the introduction of hydrophilic component (PEG) on the particle must precede their dispersion in fluorinated solvents, and only a limited amount of PEG could be introduced for good dispersibility. We were able to overcome such obstacle using the second approach, which is by introducing PEG separately into the dispersed phase and using F-SiO2 NPs in the continuous phase. The PEG coating/adsorption process took place after the particles adsorbed at the wateroil interface (to form “PEGads-F-SiO2 NPs”). The PEG density at the interface was controlled


11


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 24

independently by varying the concentration of PEG in the dispersed phase. The high solubility of PEG in aqueous solutions ensured sufficient coating density on particle surface to preserve enzyme activity. We found that enzymatic activity was fully restored when sufficient amount of PEG (>~ 1 mg/mL) was present in the droplets; increasing PEG concentration beyond this value did not affect the reaction rate (Fig. S6). This result was consistent with previous works on the covalent coating of PEG on solid substrates, which require the density of PEG chains to exceed a certain “threshold value” to retrieve full enzyme activity.7,28 Since PEG interacted with particles only after droplet formation, this approach did not decrease the dispersibility of the particles in the continuous phase, nor the formation and the stability of the droplets.

Leakage prevention for the accurate interrogation of enzymatic activities. In many assays such as the directed evolution of enzymes,2,4,5 only a fraction of the drops would contain enzymes with desired levels of activities. In order to quantify the variation in enzymatic activities from drop to drop accurately, it is critical that the contents of the drops—including the fluorophores used in fluorogenic substrates—do not leak and cause cross-contamination. To show that our drops did not leak, we constructed a model system with Amplex Red assay, where we mixed two populations of droplets with different enzymatic reaction rates. The different reaction rates were achieved by using different enzyme concentrations ([HRP] = 0.1 mU/mL and 5 mU/mL, respectively) in the two populations of drops, while the substrate concentration was fixed ([Amplex Red reagent] = 75 µM, [H2O2] = 1 mM). The population of drops with high HRP concentration ([HRP] = 5 mU/mL) was also labelled with fluorescein at a high concentration ([fluorescein] = 10 µM). The population with low HRP concentration ([HRP] = 0.1 mU/mL) was labelled with fluorescein at a low concentration ([fluorescein] = 2 µM). Fluorescein was chosen


12

Page 13 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


to label the droplets since it is known to be non-leaky even in drops stabilized by EAsurfactants11 (also see our experimental verification in Fig. S7). Drops that contained high concentrations of HRP and fluorescein were referred to as “positive drops”. Drops that contained low concentrations of HRP and fluorescein were referred to as “negative drops” Fig. 4 shows the results for the assay performed in drops stabilized by PEGads-F-SiO2 NPs and EA-surfactants respectively. After 4 hours of incubation, the PEGads-F-SiO2 NPs system produced two populations of droplets with distinct fluorescence intensities. The positive drops had stronger fluorescence signal than the negative drops. This result was consistent with the fact that the reaction rate was faster in drops containing high [HRP] than in drops containing low [HRP]. The concentration and intensity from the reaction product resorufin were both higher in these drops than in drops with low [HRP]. In contrast, in the surfactant system, the fluorescence signal was homogenized due to the leakage of resorufin from positive drops to negative drops. It was difficult to distinguish positive drops from negative drops based on resorufin fluorescence signal. The total fluorescence count of resorufin in the surfactant system was higher than that in the particle system. This fact indicates the leakage of Amplex Red substrate from negative drops to positive drops. At t=0, no leakage of the substrate was expected due to the constant initial substrate concentration among all droplets. After some incubation, however, the substrate molecules were consumed faster in positive drops than in negative drops leading to a substrate concentration gradient. Excess substrate molecules remaining in negative drops diffused into positive drops. They became oxidized to form resorufin and generated additional fluorescence signal.


13


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 24

CONCLUSIONS Table 1 summarizes the advantages of our work compared with other state-of-the-art droplet stabilizers: 1) Our particles are simple and economical to synthesize and characterize. They do not require extra synthesis and purification steps to graft PEG onto particles. 2) They are effective in preserving enzyme activity. 3) They are effective in preventing the leakage of small molecules. To our knowledge, this is the first work that demonstrated the ability to prevent both enzyme deactivation and molecular leakage. Combined with the biocompatibility with the attachment and growth of anchorage-dependent cells,16 our particles fulfill important criteria needed for the success of droplet assays. We believe that the Pickering system presented here offers a straightforward, flexible and economical platform for enzymatic studies in droplets, and will enable new opportunities for an increased range of biochemical assays.


14

Page 15 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 1: (a) Scheme illustrating droplet generation and reagent encapsulation. (b) Optical image of drops stabilized by 100 nm PEGads-F-SiO2 NPs collected from the droplet generator.


15


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 16 of 24

Figure 2: Fluorescence image of aqueous drops containing fluorescein-linked PEG (MW=5000, 0.04 mg/mL) that are stabilized by (a) EA-surfactant and (b) 100 nm PEGads-F-SiO2 NPs, respectively. Note the adsorbed PEG here is fluorescein-linked PEG only. (c) Line scans of fluorescence intensity profile across the droplets shown in Fig. 2a and 2b. We started and ended the scan around 10-20 µm away from the droplet boundary. Fluorescence intensity was normalized to the background continuous phase. Arrows indicate the direction of the line scan along which brightness was measured using ImageJ.


16

Page 17 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Figure 3:

Fluorescence intensity evolution of two representative enzymatic assays. (a)

Hydrolysis of fluorescein diphosphate (FDP, 20 µM) in the presence of bacteria alkaline phosphatase (BAP, 9.375 U/mL), and (b) oxidation of Amplex Red (50 µM) by excess hydrogen peroxide (1 mM) catalyzed by horseradish peroxidase (HRP, 5 mU/mL). The lines are guides to the eye only. The intensity values were normalized to an internal standard (see text).


17


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 24

Figure 4: (a) Scheme showing the composition of positive and negative droplets. (b-c) Fluorescence images of the droplet mixture containing different [HRP] after 4 hours of incubation, where droplets were stabilized by (b) 100 nm PEGads-F-SiO2 NPs and (c) EAsurfactants, respectively. All droplets contained 4 mg/mL PEG, 75 µM Amplex Red reagent, and fluorescein as a label for positive and negative drops at t=0. A few representative positive and negative drops are marked with blue and red circles respectively. (d-e) Histograms showing resorufin fluorescence intensity distributions of the droplet mixture after 4 hours of incubation. The intensity values were normalized to an internal standard (fluorescein at 10 µM in positive drops).


18

Page 19 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


Table 1: Comparison of this work with existing droplet stabilizers. Simplicity of synthesis

Estimated Cost ($/mL)a

Enzyme Activity

Accuracy (no leakage)

EA-surfactant

Complicated

~$ 18

High

Low

Electrostatically Modified surfactant10

Simple

~$ 0.16

High

Lowb

F-SiO2 NPs

Simple

~$ 0.9

Low

High

PEGads-F-SiO2 NPs (this work)

Simple

~$ 0.9

High

High

a

denotes cost (in USD) per mL of 2% (w/w) surfactant solution or nanoparticle suspension in

HFE-7500. See SI for details of estimation. b

See supporting information Fig. S9.


19


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 20 of 24

ASSOCIATED CONTENT Supporting Information. Detailed protocols of enzymatic assays, supplementary experimental results on droplet generation, stability and leakage studies. Cost estimation and analysis. This material is available free of charge via the Internet at http://pubs.acs.org. AUTHOR INFORMATION Corresponding Author *Email: [email protected]. Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ACKNOWLEDGMENT We acknowledge funding from the Stanford Nano Shared Facilities Bio/Medical Mini Seed Grant, Stanford CHeM-H, Stanford Bio-X, and the Stanford Woods Institute for the Environment. S.T. acknowledges additional support from 3M Untenured Faculty Award.


20

Page 21 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


REFERENCES (1) Kiss, M. M.; Ortoleva-Donnelly, L.; Beer, N. R.; Warner, J.; Bailey, C. G.; Colston, B. W.; Rothberg, J. M.; Link, D. R.; Leamon, J. H. Anal. Chem. 2008, 80, 8975-8981. (2) Agresti, J. J.; Antipov, E.; Abate, A. R.; Ahn, K.; Rowat, A. C.; Baret, J.-C.; Marquez, M.; Klibanov, A. M.; Griffiths, A. D.; Weitz, D. A. Proc. Natl. Acad. Sci. U. S. A. 2010, 107, 40044009. (3) Rosenfeld, L.; Lin, T.; Derda, R.; Tang, S. K. Y. Microfluid. Nanofluid. 2014, 16, 921-939. (4) Fischlechner, M.; Schaerli, Y.; Mohamed, M. F.; Patil, S.; Abell, C.; Hollfelder, F. Nat. Chem. 2014, 6, 791-796. (5) Zinchenko, A.; Devenish, S. R. A.; Kintses, B.; Colin, P. Y.; Fischechner, M.; Hollfelder, F. Anal. Chem. 2014, 86, 2526-2533. (6) Holtze, C.; Rowat, A. C.; Agresti, J. J.; Hutchison, J. B.; Angile, F. E.; Schmitz, C. H. J.; Koster, S.; Duan, H.; Humphry, K. J.; Scanga, R. A.; Johnson, J. S.; Pisignano, D.; Weitz, D. A. Lab Chip 2008, 8, 1632-1639. (7) Jeon, S. I.; Lee, J. H.; Andrade, J. D.; Degennes, P. G. J. Colloid Interface Sci. 1991, 142, 149-158. (8) Amiji, M.; Park, K. Biomaterials 1992, 13, 682-692. (9) McPherson, T.; Kidane, A.; Szleifer, I.; Park, K. Langmuir 1998, 14, 176-186. (10) DeJournette, C. J.; Kim, J.; Medlen, H.; Li, X.; Vincent, L. J.; Easley, C. J. Anal. Chem. 2013, 85, 10556-10564.


21


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 22 of 24

(11) Skhiri, Y.; Gruner, P.; Semin, B.; Brosseau, Q.; Pekin, D.; Mazutis, L.; Goust, V.; Kleinschmidt, F.; El Harrak, A.; Hutchison, J. B.; Mayot, E.; Bartolo, J.-F.; Griffiths, A. D.; Taly, V.; Baret, J.-C. Soft Matter 2012, 8, 10618-10627. (12) Woronoff, G.; El Harrak, A.; Mayot, E.; Schicke, O.; Miller, O. J.; Soumillion, P.; Griffiths, A. D.; Ryckelynck, M. Anal. Chem. 2011, 83, 2852-2857. (13) Sandoz, P. A.; Chung, A. J.; Weaver, W. M.; Di Carlo, D. Langmuir 2014, 30, 6637-6643. (14) Courtois, F.; Olguin, L. F.; Whyte, G.; Theberge, A. B.; Huck, W. T. S.; Hollfelder, F.; Abell, C. Anal. Chem. 2009, 81, 3008-3016. (15) Chen, Y.; Gani, A. W.; Tang, S. K. Y. Lab Chip 2012, 12, 5093-5103. (16) Pan, M.; Rosenfeld, L.; Kim, M.; Xu, M.; Lin, E.; Derda, R.; Tang, S. K. Y. ACS Appl. Mater. Interfaces 2014, 6, 21446-21453. (17) Vertegel, A. A.; Siegel, R. W.; Dordick, J. S. Langmuir 2004, 20, 6800-6807. (18) Nakanishi, K.; Sakiyama, T.; Imamura, K. J. Biosci. Bioeng. 2001, 91, 233-244. (19) Bower, C. K.; Sananikone, S.; Bothwell, M. K.; McGuire, J. Biotechnol. Bioeng. 1999, 64, 373-376. (20) Kondo, A.; Murakami, F.; Kawagoe, M.; Higashitani, K. Appl. Microbiol. Biotechnol. 1993, 39, 726-731. (21) Czeslik, C.; Winter, R. Phys. Chem. Chem. Phys. 2001, 3, 235-239. (22) Anna, S. L.; Bontoux, N.; Stone, H. A. Appl. Phys. Lett. 2003, 82, 364-366.


22

Page 23 of 24

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


(23) Anderson, R. A.; Vallee, B. L. Proc. Natl. Acad. Sci. U. S. A. 1975, 72, 394-397. (24) Welinder, K. G. Eur. J. Biochem. 1979, 96, 483-502. (25) Parida, S. K.; Dash, S.; Patel, S.; Mishra, B. K. Adv. Colloid Interface Sci. 2006, 121, 77110. (26) Derkaoui, N.; Said, S.; Grohens, Y.; Olier, R.; Privat, M. Langmuir 2007, 23, 6631-6637. (27) Preari, M.; Spinde, K.; Lazic, J.; Brunner, E.; Demadis, K. D. J. Am. Chem. Soc. 2014, 136, 4236-4244. (28) Jeon, S. I.; Andrade, J. D. J. Colloid Interface Sci. 1991, 142, 159-166. (29) Kenausis, G. L.; Voros, J.; Elbert, D. L.; Huang, N. P.; Hofer, R.; Ruiz-Taylor, L.; Textor, M.; Hubbell, J. A.; Spencer, N. D. J. Phys. Chem. B 2000, 104, 3298-3309. (30) Bergstrom, K.; Holmberg, K.; Safranj, A.; Hoffman, A. S.; Edgell, M. J.; Kozlowski, A.; Hovanes, B. A.; Harris, J. M. J. Biomed. Mater. Res. 1992, 26, 779-790. (31) Jo, S.; Park, K. Biomaterials 2000, 21, 605-616. (32) Ulbricht, M.; Matuschewski, H.; Oechel, A.; Hicke, H. G. J. Membr. Sci. 1996, 115, 31-47. (33) Peracchia, M. T.; Harnisch, S.; Pinto-Alphandary, H.; Gulik, A.; Dedieu, J. C.; Desmaele, D.; d'Angelo, J.; Muller, R. H.; Couvreur, P. Biomaterials 1999, 20, 1269-1275.


23


1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 24 of 24

For TOC only


24

Fluorinated Pickering Emulsions with Nonadsorbing Interfaces for

Recommend Documents