Patent Application
20220008880
The present invention relates to a surfactant-stabilized fluid interface, to a dispersion comprising one or more droplets having a surfactant stabilized fluid interface and to a separation process using a surfactant-stabilized fluid interface. The present invention further relates to using the interface or the dispersion comprising the same in chemical reactions.
Physical compartments in the form of emulsion droplets provide an easy to control and self-contained microenvironment. Therefore, droplet-based microfluidics has emerged as a powerful tool to produce such compartments for single cell and deoxyribonucleic acid (DNA) analysis, chemical synthesis, drug delivery as well as synthetic cell assembly. Many of these applications depend on strategies to functionalize the droplets' inner periphery in order to adjust interactions with the droplet content. This has been achieved by covalent modification of the surfactant molecules themselves, for instance with gold nanoparticles, amine groups or carboxyl groups. However, this method is not universally applicable—it depends on the success of the chemical synthesis and can interfere with the stability and the physical properties of the droplets. Moreover, the covalent binding of the functional group to the droplet periphery is irreversible.
DNA nanotechnology on the other hand can attain the programmable assembly of arbitrary nanoscale architectures like DNA-based lattices, nanopores or lid-containing boxes. DNA has also been used as a scaffold or linker to assemble secondary components including proteins, gold nanoparticles and liposomes. In addition, networks of emulsion droplets or colloid-coated droplets have been created using DNA linkers. Yet in all cases, the linkage was based on biotinylated DNA, which requires additional efforts to graft streptavidin onto the droplet surface so that the biotinylated DNA can be linked via the biotin-streptavidin interaction to the droplet surface. Furthermore, it has never been demonstrated that it is possible to functionalize the interior of water-in-oil emulsion droplets with DNA. Moreover, it was observed that synthetic mimics of cellular compartments based on lipid-based protocells lack the mechanical and chemical stability to allow their use under various conditions required for performing for example chemical reactions.
Accordingly, there is a need for a fluid interface which can be functionalized more easily and which can be used under versatile conditions.
In accordance with the present invention, this object is solved by a surfactant-stabilized fluid interface according to claim 1.
The surfactant-stabilized fluid interface according to the invention comprises a layer of surfactant and a first compound containing a hydrophobic part covalently linked to a molecular recognition site, wherein the surfactant-stabilized fluid interface has a first fluid on one side and a second fluid on the other side, wherein the surfactant stabilizes the fluid interface, wherein the hydrophobic part of the first compound is interacting with the layer of surfactant by secondary non-covalent interactions and the molecular recognition site of the first compound extends from the layer of surfactant, and wherein the surfactant is a block-copolymer having at least one hydrophilic block and at least one hydrophobic block.
This solution bases on the surprising finding that by using a block-copolymer having at least one hydrophilic block and at least one hydrophobic block as surfactant, a stable surfactant-stabilized fluid interface can be formed in which a first compound having a hydrophobic part and a molecular recognition site self assembles into the layer of surfactant so that the hydrophobic part is anchored in the layer of surfactant and the molecular recognition site extends from the surfactant. This way, the properties of the molecular recognition site can be used in versatile ways, for example for controlling the microfluidic behavior of a microfluidic droplet encapsulated by such a layer of surfactant or for catalyzing chemical reactions at the fluid interface. Since the first part is provided with a molecular recognition site it is possible to reversibly bind another molecular recognition site of a second compound. By reversibly binding another compound to the molecular recognition site of the first compound, it is possible to utilize that further compound inside or outside of a droplet encapsulated by such a fluid interface, i.e. on the side to which the hydrophilic part extends from the layer of surfactant. As a result it is possible to introduce various functional moieties to the surfactant-stabilized fluid interface so that the properties of the surfactant-stabilized fluid interface or of a droplet encapsulated by such a surfactant-stabilized fluid interface can be tailored to specific applications.
The present invention further concerns a dispersion comprising one or more droplets having a surfactant stabilized fluid interface as described in the claims and below. In the dispersion, the second fluid is the continuous phase and the first fluid is the dispersed phase, i.e. is contained in the droplets.
Moreover, the present invention also relates to use the surfactant-stabilized fluid interface of an emulsion droplet as described herein to control the interactions between the droplet interface and its content, to control chemical reactions, for single-cell analysis and droplet-based sequencing. The invention is further applicable to planar surfactant-stabilized fluid interfaces in general for surface-enhanced catalysis, medical testing, filtering or component separation.
In addition, the present invention concerns a separating process as described below and in the claims.
Preferred embodiments of the present invention are the subject-matter of dependent claims and explained in the following. It is to be understood that wherever there is no logical contradiction preferred features described below or in the dependent claims can be combined with each other.
It is preferred that the surfactant stabilized fluid interface encapsulates an emulsion droplet having a center comprising the first fluid. Even more preferably, the emulsion droplet is a microfluidic droplet.
It is particularly preferred that in the surfactant-stabilized fluid interface according to the invention the first compound has a molecular recognition site which is hydrophilic. With the molecular recognition site being hydrophilic it is easier that it extends away from the layer of surfactant. While the hydrophobic part of the first compound can be covalently bonded directly to the molecular recognition site, it is also possible that a linker is provided between the hydrophobic part and the molecular recognition site. The person skilled in the art understands that it's possible to provide the first compound with further structural elements which are not part of the hydrophobic part or of the molecular recognition site. However, in the present invention is also preferred to use first compounds consisting of a hydrophobic part, a molecular recognition site and optionally a linker between the hydrophobic part and the molecular recognition site.
The diameter of the droplet is not particularly limited. It is possible to produce droplets having a diameter of 0.1 μm or more or of 0.5 μm or more. However, with the standard methods of droplet generation by microfluidic or shaking it can be difficult to produce droplets having a diameter of less than 1 μm, it is preferred that the diameter of the droplets is 1 μm or greater, more preferred 5 μm or greater. Droplets with a diameter of 1 μm or greater can be produced for example by shaking an aqueous phase including the first compound and a non-aqueous phase including the surfactant. It is further possible to produce droplets with a diameter of 1 μm or greater by the method described below. The present invention is not limited in regard to the upper limit of upper diameter of the microdroplet. However, for more easily manufacturing the microdroplets it is preferred that the diameter of the microdroplets is not more than 500 μm. Microdroplets having a diameter of not more than 500 μm can be manufactured by using microfluidic techniques.
Particularly preferred sizes for droplets having a surfactant-stabilized fluid interface are in the range of 0.5 μm to 1,000 μm, in the range of 0.5 to 1,000 μm, in the range of 10 to 1200 μm or in the range of of 20 to 60 μm.
It is further preferred, that in the droplet encapsulated by a surfactant-stabilized fluid interface according to the invention the hydrophobic part of the first compound has a structural element based on one of fluorinated polyether, cholesterol, tocopherol, ibuprofen, palmitate, stearate, trolox or porphyrine, hydrophobic aminoacids or proteins, derivatives of any of these compounds, or other hydrophobic moieties which can be coupled to a molecular recognition site. Such hydrophobic parts are capable of inserting into artificial membranes formed by block copolymer surfactants and thereby anchoring the first compound to the surfactant layer.
Good results are particularly obtained, if the surfactant is a block copolymer comprising a lipophilic or a hydrophobic block and a hydrophilic block. The lipophilic or hydrophobic block may be, but is not restricted to members, e.g. selected from the group consisting of perfluorinated polymers, such as perfluorinated polyethers, polystyrene or poly(olefin oxides), such as poly(propylene oxide), whereas the hydrophilic block may be selected e.g. from polyether glycols, polyetheramine, polyacrylate acid, polymethylacrylate acid or poly[poly(ethylene glycol) methyl ether methacrylate].
Likewise, good results are obtained, if the polymer shell of the droplet is made of a triblock copolymer comprising two hydrophobic perfluorinated polymer end blocks and therebetween a hydrophilic polyether glycol block, wherein the triblock copolymer is folded so that the hydrophobic perfluorinated polymer blocks are arranged at the outer side and that the hydrophilic polyether glycol block is arranged at the inner side of the polymer shell. Examples for the lipophilic or hydrophobic blocks and the hydrophilic blocks are the same as those mentioned above.
Preferably, the perfluorinated polymer block is a perfluorinated polyether block (PFPE) and more preferably a perfluorinated polyether block having a weight average molecular weight of 1,000 to 10,000 g/mol. Likewise preferably, the polyether glycol (PEG) and polyetheramine (JEFFAMINE) blocks have preferably a weight average molecular weight of 100 to 10,000 g/mol. More specifically, suitable examples for the respective copolymers are PFPE-carboxylic acid (Krytox, MW 2500 or 7000 g/mol) and suitable examples for the respective diblock copolymers are PFPE (7000 g/mol)-PEG (1400 g/mol), PFPE (7000 g/mol)-PEG (600 g/mol), PFPE (2500 g/mol)-PEG (600 g/mol), PFPE (4000 g/mol)-PEG (600 g/mol), PFPE (4000 g/mol)-PEG (1400 g/mol), PFPE (2000 g/mol)-PEG (600 g/mol), PFPE (7000 g/mol)-JEFFAMINE (600 g/mol), PFPE (7000 g/mol)-JEFFAMINE (900 g/mol), PFPE (2500 g/mol)-JEFFAMINE (600 g/mol), PFPE (2500 g/mol) JEFFAMINE (900 g/mol), PFPE (4000 g/mol)-JEFFAMINE (900 g/mol), PFPE (2500 g/mol)-JEFFAMINE (600 g/mol), PFPE (2000 g/mol)-JEFFAMINE (600 g/mol), PFPE (2000 g/mol)-JEFFAMINE (900 g/mol) and suitable examples for the respective triblock copolymers are PFPE (7000 g/mol)-PEG (1400 g/mol)-PFPE (7000 g/mol), PFPE (7000 g/mol)-PEG (600 g/mol)-PFPE (7000 g/mol), PFPE (4000 g/mol)-PEG (1400 g/mol)-PFPE (4000 g/mol) PFPE (2500 g/mol)-PEG (600 g/mol) PFPE (2500 g/mol), PFPE (2000 g/mol)-PEG (600 g/mol)-PFPE (2000 g/mol), PFPE (7000 g/mol)-JEFFAMINE (900 g/mol)-PFPE (7000 g/mol) PFPE (7000 g/mol)-JEFFAMINE (600 g/mol)-PFPE (7000 g/mol), PFPE (4000 g/mol)-JEFFAMINE (900 g/mol)-PFPE (4000 g/mol), PFPE (4000 g/mol)-JEFFAMINE (600 g/mol) PFPE (4000 g/mol), PFPE (2500 g/mol)-JEFFAMINE (900 g/mol)-PFPE (2500 g/mol), PFPE (2500 g/mol)-JEFFAMINE (600 g/mol)-PFPE (2500 g/mol), PFPE (2000 g/mol)-JEFFAMINE (900 g/mol)-PFPE (2000 g/mol) and PFPE (2000 g/mol)-JEFFAMINE (600 g/mol)-PFPE (2000 g/mol). The molecular weight is determined with gel permeation chromatography using a polystyrene standard.
It is further preferred that at least one of the hydrophobic blocks of the surfactant is fluorinated. A stable droplet can be obtained, when the surfactant is a block copolymer comprising one block having a polyethylene glycol segment and comprising another block which is fluorinated. According to a particularly preferred embodiment of the present invention, the block copolymer is a diblock copolymer or a triblock copolymer. Particularly preferred surfactants are of Perflouro-polyether-polyethylene glycol (PFPE-PEG) block-copolymer fluorosurfactants, for example PFPE-PEG diblock-copolymer surfactant or a PFPE-PEG-PFPE triblock-copolymer surfactant.
According to the present invention it is preferred that one of the first and second fluids is water-based one of the first and second fluid is oil-based, e.g. the first fluid is water-based and the second fluid is oil-based, or the first fluid is oil-based and the second fluid is water-based, or the first and second fluid are water-based, or the first and second fluid are oil-based. In this regard it is particularly preferred that the first fluid is water-based, especially when the molecular recognition site of the first compound is hydrophilic. With the first fluid being water-based and the molecular recognition site of the first compound being hydrophilic, it is possible to have the molecular recognition site of the first compound extending from the layer of surfactant into the center of the droplet. This way, it is possible to use the molecular recognition site of the first compound to control release and take up of the second compound inside the controlled environment of the center of the droplet.
It is particularly preferred that the droplet according to the present invention contains a second compound comprising a further molecular recognition site which is capable of selectively binding to the molecular recognition site of the first compound. By providing such a second compound it is possible to introduce functional groups to that side of the surfactant-stabilized interface according to the invention, from which the molecular recognition site of the first compound extends away, for example to a center of a droplet encapsulated by the surfactant-stabilized interface, wherein the functional groups self-assemble to the layer of surfactant by molecular recognition. In this regard it is particularly preferred that the introduced functional groups do not bind covalently to the layer of surfactant, i.e. the second compound interacts with the first compound via secondary interactions. Accordingly, the second compound self assembles with the molecular recognition site of the first compound which extends from the layer of surfactant. This self-assembly is preferably based on secondary interactions, i.e. not on covalent bonds, so that at low temperatures, for example room temperature, the first compound and the second compound hybridize and thereby form an aggregate, for example a dimer. When increasing the temperature above room temperature, for example to temperatures of greater than 50° C., the aggregate of the first compound and of the second compound can melt and the second compound is released from the layer of surfactant. It is to be understood that the temperature at which the second compound is released from the first compound depends on the nature of the molecular recognition site and can be tuned according to need, by choosing the molecular recognition site appropriately. If the molecular recognition is for example based on formation of Watson-Crick base pairs between sequences of nucleobases, it is possible to tune the release temperature of the second compound by choosing the sequence and number of nucleobases appropriately. If for example the sequence length is increased, the release temperature can increase. However, it is also possible to increase the release temperature for example by changing the types of nucleobases in the nucleobase sequence without increasing the length of the nucleobase sequence (e.g. increasing the content of nucleobase G and/or C).
The second compound can be introduced into the droplet when manufacturing the droplet in the first place. In this case, the first and second compound can be dissolved in the same aqueous solution or alternatively can be dissolved in separate aqueous solutions which are then combined with a non-aqueous solution and the surfactant to form the droplets either by shaking or by microfluidic techniques. Alternatively, also a microfluidic particle can be manufactured containing only the first compound and separately from that a microfluidic particle can be manufactured which contains only the second compound. When combining a dispersion of the microfluidic particle containing the first compound and a dispersion of the microfluidic particle containing the second compound, there is a possibility that the microfluidic particles coalesce and form a microfluidic particle containing the first and the second compound.
It is further possible that the droplet contains in addition to the first and second compounds a third or more compounds which also has a molecular recognition site capable of forming an aggregate, e.g. a dimer, with either the first or the second compound.
The way how molecular recognition between the first compound and the second compound occurs is not particularly limited. Due to the versatility of available motives, the ease of chemical functionalization and the programmability and reversibility of the interaction, it is preferred that molecular recognition between the first and the second compound occurs via hydrogen bonds. In this regard it is therefore preferred that in the droplet the molecular recognition site of the first compound is designed to form hydrogen bonds with the molecular recognition site of the second compound. However, also antibody-anti-gen or enzymatic interaction can form the basis for molecular recognition between the first compound and the second compound. Molecular recognition between nucleic acids is a versatile and well examined example for molecular recognition sites. In this regard, it is particularly preferred that the molecular recognition site of the first compound and of the second compound are based on nucleic acids. It is further preferred that the molecular recognition sites of the first compound and of the second compound are based on DNA, RNA, or XNA, e.g. PNA, CeNA, TNA, GNA, LNA or antisense oligonucleotides, e.g. FANA, ANA. For achieving a sufficient interaction at room temperature it is even more preferred that the molecular recognition sites of the first compound and of the second compound comprise 5 or more nucleobases, even more preferred 7 or more nucleobases and most preferred 10 or more nucleobases. In nucleic acids based on PNA dimerization of two complementary PNA strands is generally stronger than a comparable dimerization of two complementary DNA strands. The reason therefore is that the DNA backbone contains negatively charged phosphate groups which repel the two dimerized DNA strands. Such negatively charged phosphate groups are not present in PNA strands. Accordingly, in case the molecular recognition sites are based on DNA or RNA it is particularly preferred that the molecular recognition sites comprise 7 or more nuclear bases, even more preferably 10 or more nuclear bases. By using molecular recognition sites based on nucleic acids, the strength of the secondary interaction between the first compound and the second compound can be tailored by choosing the sequence and the number of the nucleobases in the nucleic acids. Exemplary molecular recognition sites are presented in the sequences according to SEQ ID No. 1 to 13.
It is particularly preferred that in the droplet according to the present invention the molecular recognition site of the first compound comprises a sequence of nucleobases and the molecular recognition site of the second compound comprises a sequence of nucleobases, wherein the sequence of the nucleobases of the molecular recognition site of the first compound is at least partially complementary to the sequence of the nucleobases of the molecular recognition site of the second compound. In this regard it is noted that the length of the sequence of nucleobases of the first compound is not necessarily identical to the length of the sequence of nuclear basis of the second compound as long as sufficient complementary interaction between the nucleobases of the first compound and the nucleobases of the second compound can occur. Nevertheless, the lengths of these sequences of nuclear basis can be identical. It is particularly preferred that in the droplet according to the present invention the molecular recognition site of the first compound is capable to form Watson-Crick base pairing with the molecular recognition site of the second compound.
For using the molecular recognition site of the first compound to introduce functional groups into the controlled environment of the center of the droplet, it is particularly preferred that the molecular recognition site of the first compound extends from the layer of surfactant into the center of the droplet. This can be achieved for example, when the fluid is a water-based fluid and the molecular recognition site of the first compound contains polar groups, which are present for example in groups capable of performing hydrogen bonds.
It is particularly preferred that the molecular recognition site of the second compound is bound to a functional moiety having a functional group and/or is bound to a catalytically active domain and/or is bound to a polymer and/or is bound to a micro- or nanoparticle and/or is bound to a living cell. The type of functional group, catalytically active domain, polymer and living cell is not particularly limited. Due to the reversible interaction between the first molecular recognition site on the first compound and the second molecular recognition site on the second compound, it is possible to release and bind the functional moiety in a controlled manner by switching off the interaction between the first molecular recognition site on the first compound and the second molecular recognition site on the second compound. It is in principle possible to trigger the interaction of the first and second compound by a plurality of external stimuli. A preferred way to influence the direction of the first and second compound is by changing the temperature. With increasing temperature, interactions such as hydrogen bonds can break, so that the second compound is released from the layer of the surfactant. This can be used for example to release a catalyst from the layer of surfactant. When the temperature drops, the hydrogen bonds can form again, so that the second compound is bonded again to delay of surfactant. This way, the functional moiety can be recovered from the first or second fluid into which it has been previously released. Apart from using a change of temperature for controlling release and uptake of the second compound from or onto the layer of surfactant, it is also possible to use e.g. pH value, light, ionic conditions, buffer conditions or the presence of ligands.
It is particularly preferred that the second fluid comprises further droplets.
It is preferred that the dispersion comprises the droplets in an amount of 1015 or less droplets per ml, preferably 1012 or less droplets per ml, more preferably 109 or less droplets per ml. The number of droplets can be determined for example by counting the droplets using an optical microscope. The amount of microfluidic droplets per volume can be adjusted by controlling the flow rates in the microfluidic device and determined by high speed imaging during microfluidic production or conventional microscopy using an observation chamber.
The present invention further concerns a separating process using a surfactant-stabilized fluid interface according to the invention for separating a component A from a first fluid, wherein the method comprises the following steps: providing the first fluid in which the component A and a first compound are dissolved or dispersed, wherein the first compound comprises a hydrophobic part and a molecular recognition site, wherein the hydrophobic part is covalently linked to the molecular recognition site, and wherein the molecular recognition site is capable of binding component A; providing the second fluid in which a surfactant is dissolved; and bringing the first and second fluid into contact with each other; wherein the second fluid is immiscible with the first fluid such that a phase separation between the first and second fluid occurs and an interface between the first and second fluid is formed when the first and second fluid contact each other, and wherein the interface between the first and second fluid is a surfactant-stabilized fluid interface as described above and in the claims.
The above separating process is applicable for the specific separation of a component A from a first fluid and from one more components B, i.e. component B can be one compound or a mixture of compounds. Due to the specific separation of a component A, the separation process can be used for example for fluid purification or component separation, as long as it is possible to couple a molecular recognition site for the component A which should be linked to a hydrophobic part. Component A is not particularly limited. It can vary in size and molecular weight and can be as small as single-atom ion, in this case the molecular recognition site could for example be a ligand having specificity of said single-atom ion (e.g. a G-quadruplex which binds potassium ions), or even as large as cells. The applications of the separating process include, but are not limited to water purification, e.g. removal of contaminants like heavy metals, filtration of body fluids, e.g. removal of pathogenic components, and cell sorting, e.g. to remove bacteria or to sort mixed cell populations according to their surface markers.
It is preferred that in the separating process according to the invention an additional step of mixing the first and second fluid is performed, to form or a dispersion as described above and in the claims. Such a dispersion contains a plurality of droplets and has a much larger surface of the surfactant-stabilized fluid interface. The mixing can be performed by stirring, shaking, e.g. vortexing, or microfluidic techniques, wherein stirring and shaking are preferred in view of simpleness.
According to a preferred aspect of the separating process, an additional step of releasing the first fluid and the component(s) B is performed. This can be achieved e.g. by adding a destabilizing surfactant to destabilize the surfactant-stabilized fluid interface formed between the first and second fluid, or with a microfluidic droplet fusion device or with an electrostatic gun. The destabilizing surfactant can be added any time after formation of the surfactant-stabilized fluid interface. However, it can be preferable to add the destabilizing surfactant after a certain time has passed so that the component A is separated from the first fluid and bound to the surfactant stabilized-fluid interface via the first compound. As destabilizing surfactant any surfactant can be used which destabilizes the surfactant-stabilized fluid interface. Fluorinated alcohols, e.g. perfluoro-1-octanol (PFO), are one example for suitable destabilizing surfactants for surfactant-stabilized fluid interfaces having a block-copolymer having at least one hydrophilic block and at least one hydrophobic block as surfactant. It is preferred that the destabilizing surfactant is shorter (lower molecular weight) than the emulsion-stabilizing surfactants, so that stabilizing surfactant is replaced.
It is therefore further preferred that in the separating process according to the present invention between mixing and adding of the destabilizing surfactant an incubation step is performed. The duration of the incubation step, i.e. the time between the formation of the surfactant-stabilized fluid interface and the addition of the destabilizing surfactant is not particularly limited. It is possible that the incubation time is 10 seconds or more, preferably 1 minute or more, even more preferably 5 minutes or more. Thereby sufficient time is provided for the hydrophobic part of the first compound to self-assemble into the surfactant stabilized fluid interface. In addition, the incubation step allows for sufficient time for compound A to bind to the molecular recognition site. Usually, such self-assembly processes are finalized in short periods of times. There is no upper time limit for the incubation step, it is from an economic point of view not feasible to perform it longer than 24 hours.
It is further preferred that in the separating process after formation of the surfactant-stabilized fluid interface one or more external stimuli for releasing component A from the molecular recognition site of the first compound is applied.
Now the invention will be described in more detail also by referring to Figures and Examples which, however, do not limit the present application.
In part g of
As illustrated in part g of
In
The separating process is applicable for any fluid purification, as long as it is possible to couple a molecular recognition site for the component of interest to a hydrophobic part. Thereby, the size and the molecular weight of the components can range from atoms to cells: Applications could include water purification (e.g. removal of contaminants like heavy metals), filtration of body fluids (e.g. removal of pathogenic components), cell sorting (e.g. removal of bacteria, sorting of mixed cell populations according to their surface markers) and the like.
Here, we present a broadly applicable method for functionalizing microfluidic droplets utilizing the hydrophobic interaction of cholesterol-tagged DNA with the droplet-stabilizing surfactant. We show that DNA handles can serve as reversible anchoring points for various components including reactive groups, DNA nanostructures, beads, proteins or even cells. The use of off-the-shelf available DNA holds considerable advantages compared to standard methods for droplet functionalization, including: the broad scope of options for site-directed chemical functionalization, the addressability and programmability due to specific base pairing as well as the reversible stimuli-responsive binding properties of DNA. Therefore, cholesterol-tagged DNA handles bear great potential for various applications in droplet-based microfluidics. An overview of DNA handles described below is provided in Table 1.
Surface-functionalization of surfactant-stabilized droplets with cholesterol-tagged DNA
Droplet-based microfluidics relies on amphiphilic polymer-based agents (surfactants) for the formation of stable water-in-oil or oil-in-water emulsion droplets. We designed a short 3′ cholesterol-tagged DNA oligomer to investigate its self-assembly into the surfactant-layer of microfluidic droplets. The 16 base-long DNA (for DNA sequences see Table 1 and sequence listing) additionally contained a covalently linked fluorophore (Cy3) for visualization purposes. Having tested that the fluorescent dye does not influence the binding properties of the DNA to the droplet periphery, an Atto488-tagged DNA oligomer without cholesterol served as a control.
To gain more insights into the interaction of the cholesterol-tagged DNA and the droplet periphery, we set out to probe the mobility of the DNA in the surfactant layer. To this end, we performed fluorescence recovery after photobleaching (FRAP) measurements with the water-in-oil droplets containing cholesterol-tagged DNA. A confocal plane at the bottom of the droplet was selected and a circular bleaching area (5 μm diameter) was defined. The measurement procedure is illustrated in
A representative FRAP intensity curve as a function of time is shown in
In order to verify that the single-stranded DNA handles at the droplet periphery are accessible for duplex formation, a complementary Atto488-labeled DNA sequence (without any cholesterol-tag) was supplied via the aqueous phase during droplet production. In the presence of the DNA handles (which were not fluorescently labeled, and thus are not visible in the confocal images), the cholesterol-free DNA strand was found at the droplet periphery, indicating successful duplex formation (Supporting Information,
With this system in place, we made use of the large toolbox of chemical modifications available for DNA and functionalized the droplet periphery with a variety of components. A broadly applicable modification is the attachment of commonly used functional groups. One example is a commercially available amine-terminated DNA oligo (see Materials and Methods). We used confocal fluorescence imaging to confirm the successful attachment of the amine-modified DNA (labeled with Atto488) to the unlabeled DNA handles at the droplet periphery (see
Increasing the level of complexity one step further, we went from the attachment of purified proteins to living cells. Droplets have previously been used as microreactors for cells to perform high-throughput single cell analysis. Nonetheless such applications often require controlled interactions between encapsulated cells and the droplet periphery. We therefore set out to test if our DNA-handle-system is capable of anchoring living cells at the droplet periphery. We first incubated Jurkat cells with a cholesterol-tagged DNA duplex as used previously for the microspheres. The linkage via four rather than two DNA strands ensures a larger distance between the cell and the droplet periphery, leaving the cells in their native environment and avoiding unwanted interactions with the surfactant molecules. The confocal images in
For many applications, stimuli-responsive droplet functionalization is highly desirable. This requires mechanisms to turn interactions with the droplet periphery on and off upon demand. With the DNA-handle-system, we can make use of the temperature-dependency of Watson-Crick base pairing between strands of DNA to achieve stimuli-responsive reversibility of the binding.
We carried out temperature cycling experiments during which the DNA-containing droplets were placed on a temperature-controlled stage during confocal imaging. These experiments can reveal whether a) the cholesterol-tagged DNA is stable at the periphery even at elevated temperatures and b) the complementary DNA (which can carry diverse functional groups, as shown in
In the present study we demonstrated a versatile method for the functionalization of microfluidic droplets using DNA-tags. We showed that cholesterol-tagged DNA self-assembles into the periphery of surfactant-stabilized water-in-oil and oil-in-water droplets due to the hydrophobic interactions between the surfactant molecules and the cholesterol-tag at the interface between oil and water. Furthermore, by means of FRAP measurements we found that the DNA anchored in the surfactant layer is diffusive with D=(0.41±0.01) μm2s−1 and that the diffusivity depends on the surfactant concentration. Importantly, we proved that the DNA strand attached to the periphery remains fully addressable. It can act as a sequence-specific and programmable anchor for a variety of components attached to a complementary DNA strand. By creating a link between the droplet periphery and one of either amine-groups, a DNA lattice, plain microspheres, actin filaments or living cells, we showcased the versatility of our DNA handles. In principle, the system can be extended to any other functional group and component that can be linked to DNA, e.g. thiol-groups, adenylation, phosphorylation, various DNA nanostructures, gold nanoparticles, lipid vesicles or other types of proteins and cells. Conveniently, various functional groups covalently linked to DNA are commercially available. Lastly, we showed that droplet functionalization as a result of DNA duplex formation, is temperature dependent and fully reversible. The temperature at which unbinding happens is fully controllable by choosing a DNA sequence with the desirable melting point. In addition, other stimuli-responsive DNA motifs, like pH- and light-responsive elements, could be incorporated to achieve reversibility of droplet functionalization. All in all, this study shows the potential of DNA handles to specifically and reversibly functionalize surfactant-stabilized droplets for diverse applications in droplet-based microfluidics. It is to be understood that the effects demonstrated for DNA based recognition sites anchored to the periphery of microdroplets can be expected to also appear with other non-DNA based recognition sites based on nucleic acid fragments, including RNA and artificial nucleic acids, such as for example peptide nucleic acids or other molecular recognition sites, based for example on antibody-antigene recognition.
A set of random fixed-length DNA sequences was generated in MATLAB (MathWorks, Inc.) using the randseq command. The sequences were analyzed in NUPACK 39 and chosen such as to provide stable base-pairing with complementary sequences at room temperature and an overall low tendency to form secondary structures and self-dimers. The selected DNA oligos are DNA SEQ ID No. 1 to 13, listed also in Table 1, were purchased from Integrated DNA Technologies, Inc. or biomers.net GmbH. HPLC purification was performed for DNA oligos carrying modifications (Cy3, Cy5, FITC, ROX, Atto488, cholesterol-TEG, biotin or amine). The DNA oligos were diluted in Milli-Q water at a stock concentration of 100 μM, aliquoted and stored at −20° C. until use. Before the experiment, cholesterol-tagged DNA oligos were heated to 60° C. for 5 minutes to reduce aggregation. A concentration of 2 μM of each oligonucleotide was used in the aqueous solution, if not stated otherwise.
Microfluidic PDMS-based (Sylgard 184, Dow Corning, USA) devices for the formation of water-in-oil droplets were produced and assembled as described previously.4
Oil-droplets with a surrounding aqueous phase were produced manually by shaking. For this purpose, 10 μl of the surfactant-containing oil phase were added to 700 μl of the DNA-containing aqueous solution. The composition of the oil and the water phase were chosen as before. The probe was manually shaken until oil-in-water droplets formed (visible as a milky emulsion layer).
For imaging the production of water-in-oil droplets containing Cy3-labeled cholesterol-tagged or cholesterol-free DNA (10 μM), a fluorescence illuminator (HBO 100, Carl Zeiss AG, Germany), FS43HE Filter (Carl Zeiss AG, Germany) and a 40× objective (LD Plan-Neofluar 40×/0.6 Korr, Carl Zeiss AG, Germany) were used. For imaging purposes, lower flow rates (30 μlh−1 for aqueous phase and 120 μlh−1 for oil phase) were chosen.
A confocal laser scanning microscope LSM 800 (Carl Zeiss AG) was used for confocal imaging. The pinhole aperture was set to one Airy Unit and experiments were performed at room temperature. The images were acquired using a 20× (Objective Plan-Apochromat 20×/0.8 M27, Carl Zeiss AG) and a 63× immersion oil objective (Plan-Apochromat 63×/1.40 Oil DIC, Car Zeiss AG). Images were analyzed with ImageJ (NIH, brightness and contrast adjusted).
FRAP measurements were performed using a Leica SP5 confocal microscope (Leica Microsystems, Germany), equipped with an argon laser and a 63× oil-immersion objective (HCX PL APO 63×/1.40-0.60; Leica Microsystems GmbH). Surfactant-stabilized water-in-oil droplets functionalized with compound #1 in Table 1 containing single-stranded DNA with SEQ ID No. 1 with an Atto488-tag on the 5′ end and a cholesterol-tag on the 3′ end were sealed in an observation chamber. Subsequently, a bleaching spot with a radius of 2.5 μm was defined at the confocal plane at the bottom of the droplet. Using the FRAP-WIZZARD, five images were recorded before bleaching, followed by 6.5 s bleaching and the acquisition of 35 images after bleaching. Representative images are shown in
For all attachment experiments, confocal imaging was carried out as described before.
Amine groups: A DNA oligo with an amine-group at the 3′ end and an Atto488-label at the 5′ end form compound #3 in Table 1, containing DNA having SEQ ID No. 3 was purchased from Biomers and corresponds to the second compound 34 mentioned above. It was mixed with complementary 3′-cholesterol-tagged DNA having SEQ ID No. 1 (corresponding to the first compound 16 mentioned above) at an equimolar concentration of 2 μM in the standard buffer (Buffer 1) and encapsulated in microfluidic droplets as described.
DNA nanostructures: To demonstrate the attachment of DNA nanostructures to the droplet periphery, we chose a hexagonal DNA lattice composed of 3 unique DNA sequences.29 We modified one with a 3′ cholesterol-tag to give compound #4ii having DNA with SEQ ID No. 5 (Compound #4ii, Table 1, corresponds to first compound 16 mentioned above), a second one with a 5′ Atto488-tag having DNA with SEQ ID No. 4 (compound #4i, corresponds to second compound 34 mentioned above) and left a third one without modification having DNA with SEQ ID No. 6 (corresponds to a further second compound 34) (compound #4iii, #4i to #4iii all from Biomers) (see
Polystyrene beads: Plain polychromatic polystyrene microspheres with a diameter of 2 μm were purchased from Fluoresbrite™, Polysciences, Inc. The beads were mixed with 4 μM of compound #7 and 2 μM of compound #8 and compound #9 (Table 1) at a concentration of approximately 106 μl−1 in Buffer 1 before droplet production. Filter-free devices were used to prevent the beads from blocking the channel.
Actin: Actin (Cytoskeleton, Inc.) was stored in a buffer containing 2 mM TRIS/HCl, pH 8, 0.2 mM CaCl2, 0.2 mM ATP, 0.005% NaN3 and 0.2 mM DTT, at −80° C. The actin monomers were mixed with 10 mol % biotinylated-actin (cytoskeleton) and 1 mol % of Alexa488-labeled actin (LifeTechnologies). Before encapsulation, the monomers were polymerized into 5-10 μm sized filaments using an actin polymerization buffer (2.0 mM TRIS/HCl pH 8, 20 mM MgCl2, 0.2 mM CaCl2, 0.5 mM ATP, 0.005% NaN3 and 0.2 mM DTT) to give F-actin. After one hour of incubation (at room temperature), streptavidin (for biotin-binding) was added in 5-fold excess (10 μM) to the F-actin to enable the binding of biotinylated DNA (compound #6, Table 1). After incubation, the solution was centrifuged at 106 g for 1 h and the F-actin pellet was afterwards resuspended in HEPES pH 8.2 buffer to a final actin concentration of 80 μM. The complementary biotinylated and the cholesterol-tagged DNA were supplied via a second aqueous inlet in the microfluidic device at a concentration of 8 μM, yielding a one-to-one ratio of streptavidin primed biotinylated F-actin and DNA strands. To prevent blocking, encapsulation experiments were performed using filter-free microfluidic devices.
Jurkat cells: Jurkat, Clone E6-1 (ATCC® TIB-152 TM) were cultured in RPMI-1640 (ThermoFisher Scientific) containing 10% fetal bovine serum and 1% penicillin-streptomycin in a humidified incubator at 37° C. and 5% CO2. The medium was changed every two days and cells were maintained at a cell density of 5×106 cells/ml. Prior to the experiment, cells were harvested, centrifuged and resuspended in culture media at a final concentration of 106 cells/ml. 4 μM cholesterol-tagged DNA was added to the cell suspension before encapsulation. Because of typical cell sizes of 10-15 μm a filter-free device with 80 μm wide (instead of 30 μm) channels was used, resulting in droplets with a diameter of 80 μm.
DNA containing compounds #10 and #11 were encapsulated in microfluidic droplets containing with Buffer 1. ROX was chosen as a fluorophore due to its known temperature stability.41 Droplets were sealed in an observation chamber and imaged using a Leica SP5 microscope with a 60× oil immersion objective. The temperature was increased with a temperature-controlled microscope stage (Tokai Hit ThermoPlate TP-110) from 20° C. to 60° C. and subsequently decreased in steps of 10° C., each lasting 3 minutes. This cycle was then repeated 2.5 times. The region of interest (ROI) was changed for every temperature step to avoid bleaching. Fluorescence intensities within 20 droplets were analyzed using ImageJ.
A drop shape analyzer DSA25 (Kruss GmbH, Germany) tensiometer with CCDcamera and the pendant drop method was used to measure the surface tension at the interface of surfactants and aqueous phase in presence and absence of cholesterol-tagged DNA. The Laplace-Young equation was selected as a fittingmethod.45 Oil and aqueous phase densities were set to 1.6 gcm−3 and 0.99 gm−3, respectively. For the measurements 1 mL of the aqueous phase (10 mM Tris-HCl, 1 mM EDTA, 5 mM MgCl2, pH 8) was used in a plastic cuvette. To study to which extent cholesterol-tagged DNA alters the IFT, three solutions were investigated: one DNA-free, one with 2 μM 3′ cholesterol-tagged DNA (SEQ ID No. 1) and one with 2 μM cholesterol-tagged and complementary DNA (SEQ ID No. 3)—all in the standard buffer (Buffer 1). The oil phase contained either commercial surfactant (2 wt % of Perflouro-polyether-polyethylene glycol (PFPE-PEG) block-copolymer fluorosurfactants from Ran Biotechnologies, Inc.) or custom-synthesized surfactant (2.5 mM of a triblock-copolymer PFPE-PEG-PFPE triblock-copolymer surfactant1,2) diluted in HFE-7500 (DuPont). Prior to the measurement, samples were drawn into 1 mL syringes equipped with a cannula (0.8 mm×22 mm blunt/dull) and lowered into the aqueous phase. Drops were created using the software controlled dosing unit. Droplet interfacial tension (IFT [mN m−1]) was analyzed over time. Once the IFT did not show any change, the measurement was stopped and the IFT was collected. For each surfactant/buffer experiment, three measurements were performed and the mean value as well as the standard deviation were calculated. All results are presented in Table 2.
Table 2: Interfacial tension between aqueous and oil/surfactant phase for DNA-free aqueous solution (10 mM Tris-HCl, 1 mM EDTA, 5 mM MgCl2, pH 8), or solutions where either 2 μM cholesterol-tagged DNA or 2 μM cholesterol-tagged DNA and 2 μM of its complementary strand have been added. The measurements have been carried out for both surfactants used in this study. The results indicate no dramatic change of the IFT between aqueous and oil phase. Nonetheless, a trend to lower tension can be deduced in the presence of cholesterol-tagged DNA.
For the FRAP analysis, normalized intensity values were calculated as follows:
Ipre was calculated by taking the average of the five measured intensity values before bleaching. Due to the low number of acquired images, a correction for the bleaching of the dye during this pre-bleaching phase has not been taken into account. A non-linear least-square fit was applied to the normalized intensities from the recovery phase. The fit-function was of the form:
f(t)=A(1−exp(−λl))+x0, (2)
where A and λ, are fit parameters and x0 is the time point after bleaching i.e. the start of the recovery phase. With this, following the protocol of Axelrod5 and Soumpasis,6 the diffusion coefficient can be calculated via:
where τ is the half-recovery time and r the radius of the circular bleaching spot (2.5 μm). In order to efficiently evaluate the measured data, a custom-written MATLAB script performing the above mentioned operations was used.
In order to study the behavior of cholesterol-tagged DNA within the surfactant layer, we performed FRAP measurements for a variety of surfactant concentrations. The experiments were carried out with a commercial surfactant (008-FluoroSurfactant, RAN Biotechnologies). The results can be seen in
The diffusion coefficient decreases with an increasing surfactant concentration. A nonlinear-least square fit has been applied using the formula:
D(c)=A exp(−λc), (4)
where A and λ are fitting coefficients and c the surfactant concentration. A possible interpretation of this result is that the packing of the surfactants at the droplet periphery changes with the surfactant concentration. An increasing concentration results in a denser packing at the droplet periphery and therefore a decreased diffusion coefficient of the cholesterol-tagged DNA. It also makes sense that the packing of surfactants at the periphery is limited. As the data indicate, the diffusion coefficient remains constant above a surfactant concentration of 2 wt %. This could be due to the saturation of densely packed surfactant at the droplet periphery. It is possible that the insertion of cholesterol-tagged DNA into the surfactant layer provides an experimental handle to determine the diffusive properties of surfactants.
References to the following non-patent literature have been made above, indicated by corresponding superscripted numbers:
| Number | Date | Country | Kind |
|---|---|---|---|
| 18206772.8 | Nov 2018 | EP | regional |
| Filing Document | Filing Date | Country | Kind |
|---|---|---|---|
| PCT/EP2019/066832 | 6/25/2019 | WO | 00 |
