Microfluidic Production of Uniform Microcarriers with

Feb 17, 2016 - Jeroen Leijten , Jungmok Seo , Kan Yue , Grissel Trujillo-de Santiago , Ali Tamayol , Guillermo U. Ruiz-Esparza , Su Ryon Shin , Rohola...
5 downloads 12 Views 7MB Size
Download PDF
Article pubs.acs.org/cm

Microfluidic Production of Uniform Microcarriers with Multicompartments through Phase Separation in Emulsion Drops Nam Gi Min,† Minhee Ku,‡,§ Jaemoon Yang,‡,§ and Shin-Hyun Kim*,† †

Department of Chemical and Biomolecular Engineering, KAIST, Daejeon 34141, South Korea Department of Radiology, College of Medicine, Yonsei University, Seoul 03722, South Korea § Brain Korea 21 plus Project for Medical Science, Yonsei University, Seoul 03722, South Korea ‡

S Supporting Information *

ABSTRACT: Microfluidics has provided biodegradable microcarriers for sustained release of drugs, while allowing insignificant initial bursting. Here, we further engineer the microcarriers to have multiple compartments to release distinct drugs in a more controlled fashion. With a capillary microfluidic device, monodisperse emulsion drops containing two immiscible biocompatible polymers, one is biodegradable and the other is pH-responsive, are prepared, which undergo phase separation as the solvent in the drops is depleted. The emulsion drops finally form uniform microcarriers with distinct compartments. During the phase separation and consolidation, two model drugs with different hydrophobicity involved in the drops are spontaneously concentrated in their own compartments with higher affinity. The configuration of the microcarriers is exclusively selected from five different structures depending on the pair of polymers, pH of continuous phase, and organic solvent of emulsion drops; two of five have minimum interfacial energy, and the others are kinetically arrested. Microcarriers with each configuration provide their unique release behavior of the model drugs. These are potentially useful for choosing the release profile of multiple drugs suitable for specific diseases.

■

technique, respectively.22,23 However, core−shell geometry that is good for encapsulation and controlled release is difficult to achieve using the techniques. A simple and versatile way to create multiple domains in a single carrier is evaporationinduced phase separation of polymers in emulsion drops.24−28 For example, the organic solution of two different biodegradable polymers is microfluidically emulsified in water, which form uniform microcarriers with dual domains after drop consolidation.29−31 However, the resulting microcarriers show only a few typical configurations with poorly controlled morphology. In addition, carrier materials are restricted to biodegradable polymers which are only appealing for sustained release, restricting applicable area. More importantly, encapsulation of distinct drugs in their own domains and their release behavior are yet to be exploited. In this paper, we report a pragmatic microfluidic method to create monodisperse microcarriers with multiple domains which are able to encapsulate two different materials in their own domains and to release them in a controlled fashion. With a capillary microfluidic device, the organic solution containing biodegradable polymer and pH-responsive polymer is emulsified in water to form highly monodisperse emulsion drops. Drops evolve to microcarriers with multiple domains through phase separation as the organic solvent becomes depleted.

INTRODUCTION Programmed release of multiple ingredients or drugs at a controlled rate is of significant importance to improve therapeutic efficacy.1−5 Various types of drug carriers have been designed to fulfill the release functions.6−9 For example, uniform biodegradable microcarriers with spherical or cylindrical shapes are prepared by drop-based microfluidics and a micromolding technique, respectively, to provide a controlled rate of drug release with minimum initial bursting;10,11 conventional emulsion templates, prepared by bulk emulsification, yield polydisperse microspheres, achieving limited control over rate of polymer degradation and drug release.12 To encapsulate multiple distinct drugs and release them in a programmed fashion, complex microcarriers such as Janus microparticles,13 capsules-in-capsules,14−16 and multicored capsules17−19 have been prepared. However, to produce such complex carriers, delicate flow control is usually required to produce complex emulsion templates and only small variation of carrier configuration is possible, thereby providing limited types of the release profile; these restrict practical usage of the carriers. Although facile methods to create complex emulsions from single emulsions have been recently developed by employing controlled phase separation,20,21 it is still challenging to produce biocompatible, multicompartment microcarriers from the emulsions due to a limited set of materials available for the liquid−liquid phase separation. Alternatively, microdisks and rods with multiple compartments have been designed by stop-flow lithography and the sequential micromolding © 2016 American Chemical Society

Received: December 11, 2015 Revised: February 16, 2016 Published: February 17, 2016 1430

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials

the first drop generation for pH 8.5. Two fluorescent dyes of Nile red and Coumarin are additionally dissolved in the emulsion drops as model drugs. The emulsion drops evolve to monodisperse microcarriers, of which a unique structure is exclusively selected from five distinct configurations as illustrated in Figure 1; three of them are difficult to create in

During the phase separation, chemicals included in the drop spontaneously migrate into domains with higher affinity. Therefore, two different chemicals can be separately concentrated in their own domains. The configuration of the microcarriers is exclusively selected from several different shapes, including core−shell, acorn, inverted core−shell, coredouble shells, and multicores-in-matrix, according to a pair of polymers, organic solvent, and pH of collection liquid. The carrier configuration considerably influences the release behavior of encapsulants, which potentially enables us to select a proper release profile of drugs to specific diseases.

■

RESULTS AND DISCUSSION Phase Separation of Polymers Confined in Emulsion Drops. When two immiscible polymers are diluted in the drop, they retain a single phase because random distribution of the polymers provides sufficient entropic energy gain. As the polymers are concentrated by solvent evaporation, disfavored interaction between two immiscible polymers overwhelms the entropic effects, resulting in spontaneous phase separation to minimize Gibbs free energy of mixing, ΔGm; ΔGm has two minima, and two phases have compositions corresponding to the minima. The phase separation is initiated by spinodal decomposition by which any small fluctuation in composition is amplified by diffusion inside the spinodal region.32 As a result, the small domains of one polymer-rich phase are surrounded by domains of the other polymer-rich phase, which are then coalesced and rearranged through Ostwald ripening to reduce interfacial energy.33 Finally, two large domains survive in each drop if they have sufficient fluidity. Two immiscible fluids dispersed in a third immiscible continuous phase take one of four typical configurations at equilibrium to minimize total interfacial energy;34−40 the configurations include two drop-indrop structures with opposite compositions, drop-to-drop linkage, and two drops isolated. The two domains in the drop also take similar configurations if there is no interruption toward equilibrium state. However, the solvent evaporation involves a dynamic change of physical properties including interfacial energy and viscosity, which potentially yields a rich variety of structural configurations that consolidated polymeric microcarriers eventually take. To provide various structures of microcarriers and achieve distinct release profiles for two different drugs, we employ the phase separation of two immiscible polymers within emulsion drops. One of the polymers is biodegradable, selected from poly(lactic-co-glycolic acid) (PLGA), polycaprolactone (PCL), and poly(lactic acid) (PLA) for sustained release and the other is poly(butyl methacrylate-co-(2-dimethylaminoethyl) methacrylate-co-methyl methacrylate) (p(BMA-co-DAMA-co-MMA)) with 1:2:1 molar ratio, dissolvable in acidic conditions, used for pH-responsive release. The polymers are dissolved in organic solvent of either chloroform or toluene. The binary polymer solution is uniformly emulsified into continuous phase using a glass capillary microfluidic device as shown in Figure S1. As a continuous phase, we use an aqueous solution of 10 w/w% poly(vinyl alcohol) (PVA) and 0.1 M NaOH. The solution has pH of 12 as prepared, which decreases over time to 8.5 within 3 h, as shown in Figure S2. This is because PVA is hydrolyzed in the basic condition which releases protons and neutralizes the solution. To control pH to be approximately 8.5 and 10 during phase separation in drops, respectively, we collect drops for a short period of time and observe them: 5 min immediately after the first drop generation for pH 10 and 10 min after 1 h from

Figure 1. Schematic illustration showing the formation of microcarriers with five distinct configurations through phase separation between biodegradable and pH-responsive polymers confined in emulsion drops.

the absence of the kinetic effect. The selection is influenced by the composition of polymers, pH value of continuous phase, and organic solvent of emulsion drops. Microcarriers with Biodegradable Core and pHResponsive Shell. The emulsion drops of chloroform solution of hydrophobic PLGA and hydrophilic p(BMA-co-DAMA-coMMA) are homogeneous as prepared at approximately pH 8.5, which then become heterogeneous as they are shrunken by chloroform diffusion to the surrounding. Within drops, very small domains of p(BMA-co-DAMA-co-MMA)-rich phases are formed in a continuous PLGA-rich phase by spinodal decomposition, which are continuously coalesced to form a large core in each drop. When the core is then brought into a contact with continuous phase, the core and shell phases are spontaneously inverted, finally resulting in the microcarriers composed of the PLGA-rich core and the p(BMA-co-DAMAco-MMA)-rich shell, as shown in Figure 2a,b and Movie S1. The phase inversion is driven by interfacial energy. Because the interfacial tension between the p(BMA-co-DAMA-co-MMA)rich and continuous phases is much lower than that between PLGA-rich and continuous phases, the p(BMA-co-DAMA-coMMA)-rich phase spreads on the surface of the PLGA-rich phase to minimize the interfacial area between the PLGA-rich and continuous phase; that is, the spreading parameter for the p(BMA-co-DAMA-co-MMA)-rich phase is positive.36,37 The core−shell geometry is maintained until drops are fully consolidated, yielding spherical core−shell microcarriers, as shown in Figure 2c and the inset of Figure 2c. The microcarriers are modeled by monodisperse emulsion drops, thereby showing narrow size distribution; average diameter is 25.70 μm and its coefficient of variation (CV) is as small as 1.58%. During the phase separation and consolidation, Nile red and Coumarin are spontaneously concentrated into PLGA-rich and p(BMA-co-DAMA-co-MMA)-rich domains, respectively. The concentration of Nile red in the PLGA-rich core is approximately 2.23 times (CV: 12.6%) higher than that in the 1431

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials

Acorn-Shaped Janus Microcarriers. The chloroform drops with the same composition, PLGA and p(BMA-coDAMA-co-MMA), dispersed in the continuous phase with pH 10 evolve to acorn-shaped Janus microcarriers rather than core−shell. Drops experience the spinodal decomposition, subsequent internal coalescence, and phase inversion until the middle stage of phase separation in the same manner to continuous phase with pH 8.5. However, the p(BMA-coDAMA-co-MMA)-rich phase dewets on the surface of the PLGA-rich phase, as shown in Figure 3a,b. The boundary

Figure 2. (a, b) Series of illustration and optical microscope (OM) images showing the formation of a microcarrier composed of a biodegradable core and pH-responsive shell from a homogeneous chloroform drop containing p(BMA-co-DAMA-co-MMA) and PLGA in aqueous phase with pH 8.5. Two different dye molecules of Nile red and Coumarin are additionally dissolved in the drop. (c) Confocal laser scanning microscope (CLSM) images of monodisperse core− shell microcarriers. (d) A set of red (left panel), green (middle panel), and overlaid (right panel) fluorescence CLSM images and intensity profile of red and green fluorescence along the dotted line in the overlaid image. (e) A series of CLSM images showing selective dissolution of the pH-responsive shell at pH 5. Figure 3. (a, b) Series of illustration and OM images showing the formation of acorn-shaped microcarrier composed of pH-responsive nut and biodegradable cupule from a homogeneous chloroform drop containing p(BMA-co-DAMA-co-MMA) and PLGA in aqueous phase with pH 10. (c) CLSM images of acorn-shaped microcarriers with uniform size and shape. (d) A set of red (left panel), green (middle panel), and overlaid (right panel) fluorescence CLSM images and intensity profile of red and green fluorescence along the dotted line in the overlaid image. (e) A series of CLSM images showing selective dissolution of pH-responsive nut at pH 5.

p(BMA-co-DAMA-co-MMA)-rich shell, whereas the concentration of Coumarin in the shell is approximately 1.79 times (CV: 20.5%) higher than that in the core, as shown in Figure 2d. This spontaneous, partial separation is attributed to opposite molecular affinity of dyes to polymers. Coumarin is relatively hydrophilic and has higher affinity to p(BMA-coDAMA-co-MMA) than PLGA; that is, Coumarin has a smaller interaction parameter with p(BMA-co-DAMA-co-MMA) than PLGA, which results in spontaneous migration of Coumarin toward p(BMA-co-DAMA-co-MMA)-rich domains to minimize Gibbs free energy of mixing. In contrast, hydrophobic Nile red has a smaller interaction parameter with PLGA than p(BMA-coDAMA-co-MMA), resulting in migration to PLGA-rich phase.21 Therefore, drug molecules with different hydrophobicity can also be potentially encapsulated into their own domains. The core−shell microcarriers are stable against dissolution in neutral pH condition, thereby retaining the dyes. When the microcarriers are exposed to the acidic condition of pH 5, the shell of p(BMA-co-DAMA-co-MMA) becomes selectively dissolved within 10 s, releasing the green dye to the surrounding as shown in Figure 2e and Movie S2; release by pH- or specific enzyme-responsive dissolution of one compartment in Janus microparticles has been reported.13,40 Tertiary amines of DAMA groups in the polymer are ionized to be cationic in the acidic condition, being soluble in water. Biodegradable cores remain undissolved as shown in Figure S3; in general, the PLGA microsphere degrades in a time scale of a month.41

between two polymeric phases is convex for the PLGA-rich core in the beginning of the dewetting of the shell, which becomes flat and then concave during the consolidation, ending up with acorn-shaped Janus microcarriers composed of p(BMA-co-DAMA-co-MMA)-rich nut and PLGA-rich cupule, as shown in Figure 3c and the inset of Figure 3c. Nile red and Coumarin are concentrated into PLGA-rich and p(BMA-coDAMA-co-MMA)-rich phases, respectively, in the same manner. Concentration of Nile red in the PLGA-rich cupule is approximately 3.15 times (CV: 12.2%) higher than that in the p(BMA-co-DAMA-co-MMA)-rich nut, whereas concentration of Coumarin in the nut is approximately 2.72 times (CV: 21.4%) higher than that in the cupule, as shown in Figure 3d. This evolution from the core−shell structure to the acornshaped structure with PLGA-rich cupule is attributed to large interfacial tension between p(BMA-co-DAMA-co-MMA)-rich and continuous phases at pH 10. DAMA groups in p(BMA-coDAMA-co-MMA) are partially ionized at pH 8.5 which yields 1432

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials very small interfacial tension with the continuous phase, thereby leading to positive spreading parameter and complete core−shell structure. In contrast, the ionization is highly suppressed at pH 10 which increases the tension. As a result, spreading parameter becomes negative, leading to dewetting of the shell on the core.29−31 Moreover, the tension becomes even larger than that between the PLGA-rich and continuous phases, producing the acorn-shaped Janus microcarriers with PLGArich cupule. The acorn-shaped Janus microcarriers also show selective dissolution of p(BMA-co-DAMA-co-MMA)-rich domain and release of Coumarin when they are subjected to acidic condition as shown in Figure 3e and Movie S3. The release is relatively anisotropic due to the geometry of microcarriers. After the dissolution, PLGA-rich domains with crescent-moon shape remain. Inverted Core−Shell Microcarriers. When PCL is used as a biodegradable polymer, instead of PLGA, while maintaining the other conditions including the solvent of chloroform, the final structures become phase-inverted core−shell at pH 8.5; core is p(BMA-co-DAMA-co-MMA)-rich phase and shell is PCL-rich phase. Upon spinodal decomposition, small dispersed domains of p(BMA-co-DAMA-co-MMA)-rich phase are formed in a similar manner to the combinations of PLGA and p(BMAco-DAMA-co-MMA). The domains coalesce to form a large core, and finally, microcarriers with p(BMA-co-DAMA-coMMA)-rich core and PCL-rich shell are prepared, as shown in Figure 4a−c; Coumarin and Nile red stain the core and shell, respectively. The concentration of Nile red in the PLGA-rich

shell is approximately 1.86 times (CV: 2.37%) higher than that in the p(BMA-co-DAMA-co-MMA)-rich core, whereas the concentration of Coumarin in the core is approximately 2.58 times (CV: 11.2%) higher than that in the shell, as shown in Figure 4d. We believe that inverted core−shell structures are thermodynamically unstable due to higher interfacial tension between PCL-rich and continuous phases than that between p(BMA-co-DAMA-co-MMA)-rich and continuous phases; that is, the spreading parameter for the p(BMA-co-DAMA-coMMA)-rich phase is positive. Nevertheless, phase separation leads to formation of the p(BMA-co-DAMA-co-MMA)-rich dispersed phase in the PCL-rich surrounding, which is kinetically arrested by fast consolidation before experiencing the phase inversion; we observe the phase inversion at pH 10. The fast consolidation is attributed to low PVA coverage at the interface. PVA becomes negatively charged by releasing protons to reduce pH from 12 to 8.5. Negatively charged PVA is unable to stabilize the oil−water interface, possibly yielding low interfacial coverage. The inverted core−shell structures exhibit very slow dissolution of a p(BMA-co-DAMA-co-MMA)-rich core in the acidic condition as shown in Figure 4e and Movie S4. This is because the core is protected by the biodegradable PCL shell. The intact shell allows limited diffusion of proton and prevents the release of Coumarin to the surrounding. As the shell is partially degraded, the core releases Coumarin. After the release from the core, the shell persists for a longer period of time, slowly releasing Nile red. When the shell is incomplete due to formation of additional small p(BMA-co-DAMA-coMMA)-rich domains, the ionized polymer becomes released from the main core through the defects. The release is relatively fast in comparison with the complete shell but still slow in comparison with plain core−shell and acorn-shaped Janus microcarriers. The polymers become swollen in the core and then extruded through the defect, forming arms on the shell surface, as shown in Figure S4. The release of the polymer leaves behind water-filled core in the microcarrier. Microcarriers with Single Core and Double Shells. When emulsion drops are collected in a continuous phase with pH of 10, instead of pH of 8.5, while maintaining the other conditions including the solvent of chloroform, the drops transform to microcarriers with single core and double shells; the core and the outmost shell are p(BMA-co-DAMA-coMMA)-rich, and the inner shell is PCL-rich. In the early stage of phase separation, small p(BMA-co-DAMA-co-MMA)-rich domains are formed in the PCL-rich surrounding, which then coalesce to form a large core in the same manner to pH 8.5. Afterward, the core−shell structures further undergo phase inversion before the phase separation is fully completed, forming a p(BMA-co-DAMA-co-MMA)-rich shell. Finally, the remaining small p(BMO-co-DAMA-co-MMA)-rich domains, which do not participate in the phase inversion, coalesce to form a single core as shown in Figure 5a,b. At pH 10, PVA is not fully ionized yet, being surface-active. Therefore, PVA strongly adsorbs at the oil−water interface, which retards diffusion of chloroform. This provides a sufficient time for the phase inversion. After the phase inversion, the core further experiences phase separation and coalescence of p(BMA-coDAMA-co-MMA)-rich domains, eventually forming a single p(BMA-co-DAMA-co-MMA)-rich core and double shells of the PCL-rich inner layer and p(BMA-co-DAMA-co-MMA)-rich outer layer as shown in Figure 5c,d; the concentration of Nile red in the PLGA-inner shell is approximately 1.82 (CV: 8.05%) and 1.20 times (CV: 5.95%) higher than those in the

Figure 4. (a, b) Series of illustration and OM images showing the formation of microcarrier composed of pH-responsive core and biodegradable shell from a homogeneous chloroform drop containing p(BMA-co-DAMA-co-MMA) and PCL in aqueous phase with pH 8.5. (c) CLSM images of monodisperse inverted microcarriers. (d) A set of red (left panel), green (middle panel), and overlaid (right panel) fluorescence CLSM images and intensity profile of red and green fluorescence along the dotted line in the overlaid image. (e) A series of CLSM images showing very slow dissolution of pH-responsive core at pH 5. 1433

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials

coalescence of the small domains and phase inversion during the period of consolidation, ending up with pH-responsive microcarriers with small biodegradable particles embedded, as shown in Figure 6a−c and the inset of Figure 6c. Both red and

Figure 5. (a, b) Series of illustration and OM images showing the formation of microcarrier composed of pH-responsive core, biodegradable inner shell, and pH-responsive outer shell from a homogeneous chloroform drop containing p(BMA-co-DAMA-coMMA) and PCL in aqueous phase with pH 10. (c) CLSM images of the double-shelled microcarriers. (d) A set of red (left panel), green (middle panel), and overlaid (right panel) fluorescence CLSM images and intensity profile of red and green fluorescence along the dotted line in the overlaid image. (e) A series of CLSM images showing dissolution of pH-responsive outer shell and core at pH 5.

Figure 6. (a, b) Series of illustration and OM images showing the formation of microcarrier composed of biodegradable multicores embedded in pH-responsive matrix from a homogeneous toluene drop containing p(BMA-co-DAMA-co-MMA) and PLGA in aqueous phase with pH 8.5; the same structures are obtained in pH 8.5−10. (c) CLSM images of the microcarriers with a structure of multicores-inmatrix. (d) A set of red (left panel), green (middle panel), and overlaid (right panel) fluorescence CLSM images and intensity profile of red and green fluorescence along the dotted line in the overlaid image. Both red and green fluorescence is observed along the whole volume of microcarriers. (e) A series of CLSM images showing selective dissolution of pH-responsive matrix and subsequent spreading of release of biodegradable cores at pH 5.

p(BMA-co-DAMA-co-MMA)-rich outer shell and core, whereas concentrations of Coumarin in the outer shell and core are approximately 1.68 (CV: 10.7%) and 1.67 times (CV: 9.68%) higher than that in the inner shell, respectively, as shown in Figure 5d. Although the majority has single-cored triple structure, dual- or multicored triples are also formed at a small fraction, as shown in Figure S5. The microcarriers with single core and double shells show sequential dissolution of p(BMA-co-DAMA-co-MMA)-rich domains in the acidic condition. The outer shell is rapidly dissolved within 10 s in buffer solution of pH 5, as shown in Figure 5e and Movie S5; the dissolution time scale is comparable with core−shell microcarriers in Figure 2. In contrast, the core is slowly dissolved due to partial protection with the inner PCL shell; it takes approximately 60 s. This dissolution is relatively fast in comparison with inverted core−shell microcarriers. We attribute this to formation of small p(BMA-co-DAMA-coMMA)-rich domains in the inner PCL shell due to incomplete phase separation. Microcarriers with Multicores-in-Matrix. The solvent for the dispersed phase influences the final structure of microcarriers. When toluene is used as solvent for a combination of p(BMA-co-DAMA-co-MMA) and PLGA, instead of chloroform, the emulsion drops evolve to spherical microcarriers composed of many small PLGA-rich domains embedded in a p(BMA-coDAMA-co-MMA)-rich matrix in a wide range of pH 8.5−10. In the early stage of phase separation, very small PLGA-rich domains are formed in the p(BMA-co-DAMA-co-MMA)-rich surrounding in the opposite manner to all the cases with chloroform solvent. The drops do not undergo a dramatic

green fluorescence are observed from the whole volume of microcarriers in Figure 6d. We attribute this freezing of the structure without significant coalescence to fast evaporation of toluene from drops. Toluene drops float immediately underneath the interface between air-continuous phases until toluene is almost depleted due to the low density of toluene, 0.87 g/ mL. Therefore, toluene rapidly vaporizes to air through only the thin layer of water. The consolidation of toluene drops is approximately 2.4-fold faster than that of chloroform drops which sediment on the bottom of the glass dish. When the composite microcarriers are subjected to an acidic condition of pH 5, the p(BMA-co-DAMA-co-MMA)-rich matrix is isotropically dissolved in 50 s, thereby releasing Coumarin as shown in Figure 6e and Movie S6. The dissolution leaves behind small PLGA-rich particles, which slowly spread in the continuous phase while retaining Nile red. The PLGA-rich particles are much smaller than spherical or crescent-moon-shaped compartments left behind from the core−shell or acorn-shaped microcarriers, potentially providing faster release of drugs in PLGA-rich domains. To roughly estimate the increase of surface area, we collect and observe the PLGA-rich particles with SEM as shown in Figure S6. The average diameter of the 1434

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials particles is 2.6 μm (CV: 23.0%). The surface area of the particles is increased approximately 13 times in a comparison with the single cored particle. Therefore, the microcarriers with multicores-in-matrix will provide a faster release of drugs from the multicores. When toluene is used for a combination of p(BMA-co-DAMA-co-MMA) and PCL, drops transform to microcarriers with biodegradable core and pH-responsive shell in the pH range of 8.5−10. A combination of p(BMA-coDAMA-co-MMA) and PLA in toluene produces microcarriers with parallel linkage of two domains in the pH range of 8.5−10. The only difference of configuration between toluene and chloroform for this combination of polymers is the shape of the boundary between the two domains; chloroform results in a concave boundary for the PLA-rich domain, whereas toluene gives a relatively flat boundary. All representative configurations of microcarriers obtained by sets of polymers, organic solvents, and pH values are summarized in Figure 7.

is released within 2000 s for all three pH values, whereas only half of Nile red is released. This is because the Coumarinconcentrated p(BMA-co-DAMA-co-MMA)-rich compartments are selectively dissolved at the acidic condition. A large reduction of intensity at 600 nm indicates that Nile red is also included in p(BMA-co-DAMA-co-MMA)-rich compartments. Release of the model compounds is faster in lower pH. It takes 450 s to release 80% of encapsulated Coumarin at pH 6, whereas it is 1410 s at pH 7. We study the release profile from three different geometries of microcarriers: core−shell, inverted core−shell, and multicores-in-matrix are used; acorn-shaped and double-shelled microcarriers show almost the same release behaviors as core−shell. The microcarriers are dried and immersed in a buffer solution with pH 6.5. Core−shell microcarriers release more than 90% Coumarin and half of Nile red, as we observed in Figure 8a,b. It takes 690 s to release 80% of encapsulated Coumarin. In contrast, the inverted core−shell microcarriers retain their fluorescence spectra for 1600 s without reduction, indicating negligible release; this is consistent with CLSM images in Figure 4e. Microcarriers with multicores-in-matrix show release of Coumarin in the time scale similar to that of core−shell. At the same time, fluorescence intensity at 600 nm is also decreased. This is because small PLGA-rich particles spread into the surrounding as matrix is degraded, as we confirmed in Figure 6e; Nile red is still in the small PLGA-rich particles. We expect that Nile red will be released at reduced time scale in multicores-in-matrix in comparison with simple core−shell due to large surface area, as we discussed above. Selective Release in a Cancer Cells Medium. In general, cancer cells make an acidic environment due to the metabolic production of lactic acid by active anaerobic glycolysis.42 Therefore, pH value of the tumor environment is related to the activity of the cancer cells and is used as one of indicators for malignancy of tumors. The microcarriers we designed are able to selectively release therapeutics in an acidic environment, thereby being potentially useful for cancer treatment; although vehicles smaller than 200 nm are required to deliver drugs to tumor sites through the enhanced permeation and retention (EPR) effect, we believe that our microcarriers can be directly injected into tumor sites or implanted after surgery to release drugs. To investigate release behavior for microcarriers during cultivation of cancer cells, the dispersion of microcarriers is incubated with attached MDA-MB-231 breast cancer cells for 24 h. During the incubation, pH of the medium slowly decreases to 6.72. The p(BMA-co-DAMA-co-MMA)-rich shells are gently swollen at such a weakly acidic environment, instead of being rapidly dissolved. The shell volume becomes increased almost 5-fold in 24 h, whereas the core remains unchanged, as shown in Figure 9 and Movie S7. This swelling leads to formation of large pores in the shell, allowing slow diffusion of Coumarin to the surrounding. In contrast, the core retains Nile red. For a comparison, we incubate acorn-shaped microcarriers in the solution with constant pH of 6.72 and monitor them, as shown in Movie S8. The p(BMA-co-DAMA-co-MMA)-rich compartments are selectively swollen in 40 min and dissolved in 1 h. This experiment indicates that the cultivation medium of cancer cells has higher pH than 6.72 during most periods of incubation. We attribute this very slow reduction of pH to ionization of sodium bicarbonate included in the medium; the ionization leads to the production of bicarbonate which suppresses the ionization of carbonic acid.43

Figure 7. Summary of microcarrier configurations obtained from 12 different sets of biodegradable polymer, pH of continuous phase, and organic solvent.

Release of Model Drugs in Acidic Condition. To study release behaviors of model compounds, we measure time series of fluorescence spectra from microcarriers under acidic conditions. For this, the microcarriers are dried on a glass dish, which are then filled with a large amount of pH buffer solution. The microcarriers release dyes to the surrounding and lose their fluorescence over time; the released dyes are highly diluted, which has negligible influence on the spectra. For example, core−shell microcarriers show a fluorescence spectrum with two peaks at 510 and 600 nm which are due to Coumarin and Nile red, respectively, as shown in Figure S7. The peak intensities decrease over time when the core−shell microcarriers are subjected to an acidic condition. To quantitatively compare the release of Coumarin and Nile red from microcarriers, peak intensities at 510 and 600 nm are separately plotted over time for three different pH values of 7, 6.5, and 6, as shown in Figure 8a,b. More than 90% Coumarin 1435

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials

Figure 8. (a, b) Time dependence of normalized fluorescence intensities at wavelength of 510 nm (a) and 600 nm (b) measured from core−shell microcarriers which are under denoted pH values of 7, 6.5, and 6, where emission at 510 nm is predominantly due to Coumarin and that at 600 nm is due to Nile red. (c, d) Time dependence of normalized fluorescence intensities at wavelength of 510 nm (c) and 600 nm (d) measured from the microcarriers with core−shell, inverted core−shell, and multicores-in-matrix configurations, where the pH of the surrounding is set to be 6.5.

each configuration provide their own release profile, which potentially enables us to program the release profile of multiple drugs. In particular, as pH-responsive compartments can be dissolved in the weakly acidic environment of cancers, while the biodegradable compartment is slowly degraded, the microcarriers can be designed to release two distinct cancer drugs in sequence for combinatorial chemotherapy.44 Moreover, the configuration of microcarriers and therefore release behavior can be further diversified by employing three or more different polymers in the single emulsion drops. Therefore, we believe that this general microfluidic technology will provide new opportunity for designing microcarriers with a programmed release function of multiple drugs.

Figure 9. Series of fluorescence microscope images of the core−shell microcarriers incubated with live MDA-MB-231 cells.

■

■

CONCLUSION We report a microfluidic approach to create monodisperse microcarriers composed of biodegradable and pH-responsive compartments using phase separation of two immiscible polymers confined in emulsion drops. The emulsion drops containing two polymers are homogeneous at the beginning, which then transform to microcarriers with multiple compartments as solvent is depleted in the drops. Two different model drugs of dye molecules dissolved in the emulsion drops are spontaneously concentrated into their own compartments with higher molecular affinity. Configuration of microcarriers is determined by three control factors: pair of polymers, pH value of continuous phase, and organic solvent. We can exclusively produce uniform microcarriers with unique configuration by controlling these factors from five distinct structures that are core−shell, acorn-shaped, inverted core−shell, core-double shells, and multicores-in-matrix structures. Microcarriers with

EXPERIMENTAL SECTION

Materials. Cationic copolymer, p(BMA-co-DAMA-co-MMA) with 1:2:1 molar ratio (Eudragit EPO, Mw 47 000, Evonik), is used for the pH-responsive polymer. PLGA (Mw 113 000, Evonik), PCL (Mw 45 000, Sigma-Aldrich), and PLA (Mw 25 000, Polyscience, Inc.) are used for biodegradable polymers. The cationic copolymer and one of biodegradable polymers are dissolved in either toluene (SigmaAldrich) or chloroform (Sigma-Aldrich) to have a concentration of 1 w/w% for each polymer; this organic solution is used as a dispersed phase of emulsions. Fluorescent dyes of Nile red (Sigma-Aldrich) and Coumarin (Sigma-Aldrich) are also dissolved in the organic solution with concentrations of 0.05 w/w% and 0.1 w/w%, respectively, as model drugs. As a continuous phase, an aqueous solution of 10 w/w% PVA (Mw 13 000−23 000, Sigma-Aldrich) containing 0.1 M sodium hydroxide (NaOH, Sigma-Aldrich) is used. Device Preparation and Operation. Two cylindrical capillaries (1B100F-6, World precision instruments, Inc.) are tapered by a micropipette puller (P97, Sutter Instrument), one of which is then 1436

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials sanded to have a larger orifice diameter. These two capillaries are coaxially assembled in a square capillary (OD 1.5 mm, ID 1.05 mm, Harvard borosilicate square tubing) without chemical modification. Volumetric flow rates are controlled by syringe pumps (Legato 100, KD Scientific). The flow rates of the dispersed and continuous phases are typically set to be 200 and 1500 μL h−1 and further controlled to maintain dripping mode of drop formation for different sets of fluids and devices. The production of emulsion drops is observed with an inverted optical microscope (Eclipse TS100, Nikon) equipped with a high speed camera (Phantom Miro eX2, Vision Research). Drops are collected in the continuous phase at room temperature. The solution is in a wide glass Petri dish with a height of 250 μm which facilitates evaporation of organic solvent from emulsion drops. To maintain a high pH of approximately 10 during consolidation, drops were collected the first 5 min and observed. For consolidation at approximately pH 8.5, drops are collected after 1 h and observed; the pH value continuously decreases, and we take rough mean values during drop consolidation. Microcarriers are washed three times with a basic solution and then stored in phosphate buffer solution with pH 7.4 (Sigma-Aldrich). Characterization. To observe selective dissolution of a p(BMA-coDAMA-co-MMA)-rich domain at an acidic condition, dried microcarriers are immersed in pH 5.0 buffer solution and observed with a confocal laser scanning microscope (CLSM, LSM 5 PASCAL, Zeiss). After the dissolution, remaining parts are observed with a scanning electron microscope (SEM, S-4800, Hitachi). To further confirm the selective dissolution, core−shell microcarriers are incubated in a culture medium of cancer cells, MDA-MB-231. For this, MDA-MB231 cells are plated on 6-well plate (Nunc) in densities of 2 × 105 cells/well in phenol red free RPMI1640 (GIBCO) medium supplemented with 10% fetal bovine serum and grown for 3 days in a humidified incubator containing 5% CO2 at 37 °C. Prior to live cell imaging, the cells are moved and kept on a heated microscope chamber at 37 °C in a 5% CO2-humidified atmosphere. The core− shell microcarriers are pipetted into the cell culture medium, and the cells are incubated with the core−shell microcarriers for an additional 24 h. The images are captured every 10 min by using an Automated Microscope System with Adaptive Focus Control (DMI6000B, Leica) and the LAS X software (ver. 1,1,12420,0, Leica), designed for acquisition and processing of fluorescence images (filter set: FL1, green Ex: 450−490 nm, Em: 500−550 nm; FL2, red, Ex: 565−605 nm, Em: 622−698 nm).

■

■

and subsequent spreading of PLGA-rich core particles at pH 5 (AVI) Selective swelling of p(BMA-co-DAMA-co-MMA)-rich shell of core-shell microcarriers over a course of incubation in MDA-MB-231 cancer cell medium (AVI) Selective swelling and subsequent dissolution of p(BMAco-DAMA-co-MMA)-rich compartment of acorn-shaped microcarriers in the solution with constant pH of 6.72 (AVI)

AUTHOR INFORMATION

Corresponding Author

*E-mail: [email protected]. Notes

The authors declare no competing financial interest.

■

ACKNOWLEDGMENTS This work was supported by the Midcareer Researcher Program (2014R1A2A2A01005813) and Global Research Laboratory (NRF-2015K1A1A2033054) through the National Research Foundation (NRF), funded by the Ministry of Science, ICT, and Future Planning (MSIP).

■

REFERENCES

(1) Shim, T. S.; Kim, S.-H.; Yang, S.-M. Elaborate Design Strategies Toward Novel Microcarriers for Controlled Encapsulation and Release. Part. Part. Sys. Char. 2013, 30, 9−45. (2) Shin, H.-C.; Alani, A. W.G.; Rao, D. A.; Rockich, N. C.; Kwon, G. S. Multi-drug loaded polymeric micelles for simultaneous delivery of poorly soluble anticancer drugs. J. Controlled Release 2009, 140, 294− 300. (3) Aw, M. S.; Kurian, M.; Losic, D. Polymeric Micelles for Multidrug Delivery and Combination Therapy. Chem. - Eur. J. 2013, 19, 12586− 12601. (4) Stubbe, B. G.; De Smedt, S. C.; Demeester, J. Programmed Polymeric Devices” for Pulsed Drug Delivery. Pharm. Res. 2004, 21, 1732−1740. (5) Shao, Y.; Fu, J. Integrated Micro/Nanoengineered Functional Biomaterials for Cell Mechanics and Mechanobiology: A Materials Perspective. Adv. Mater. 2014, 26, 1494−1533. (6) Win, K. Y.; Feng, S.-S. Effects of particle size and surface coating on cellular uptake of polymeric nanoparticles for oral delivery of anticancer drugs. Biomaterials 2005, 26, 2713−2722. (7) Langer, R. Tissue Engineering. Mol. Ther. 2000, 1, 12−15. (8) Duncanson, W. J.; Lin, T.; Abate, A. R.; Seiffert, S.; Shah, R. K.; Weitz, D. A. Microfluidic synthesis of advanced microparticles for encapsulation and controlled release. Lab Chip 2012, 12, 2135−2145. (9) Björnmalm, M.; Yan, Y.; Caruso, F. Engineering and evaluating drug delivery particles in microfluidic devices. J. Controlled Release 2014, 190, 139−149. (10) Xu, Q.; Hashimoto, M.; Dang, T. T.; Hoare, T.; Kohane, D. S.; Whitesides, G. M.; Langer, R.; Anderson, D. G. Preparation of Monodisperse Biodegradable Polymer Microparticles Using a Microfluidic Flow-focusing Device for Controlled Drug Delivery. Small 2009, 5, 1575−1581. (11) Acharya, G.; Shin, C. S.; Vedantham, K.; McDermott, M.; Rish, T.; Hansen, K.; Fu, Y.; Park, K. A study of drug release from homogeneous PLGA microstructures. J. Controlled Release 2010, 146, 201−206. (12) Sansdrap, P.; Moës, A. J. Influence of manufacturing parameters on the size characteristics and the release profiles of nifedipine from poly(DL-lactide-co-glycolide) microspheres. Int. J. Pharm. 1993, 98, 157−164. (13) Marquis, M.; Renard, D.; Cathala, B. Microfluidic Generation and Selective Degradation of Biopolymer-Based Janus Microbeads. Biomacromolecules 2012, 13, 1197−1203.

ASSOCIATED CONTENT

S Supporting Information *

The Supporting Information is available free of charge on the ACS Publications website at DOI: 10.1021/acs.chemmater.5b04798. OM images showing monodisperse emulsion drops, inverted core−shell microcarriers with defects on the shell, microcarriers with single-, dual-, triple-, multicores and double shells, and sets of CLSM and SEM images showing two types of microcarriers before and after dissolution. (PDF) Phase separation between PLGA and p(BMA-co-DAMAco-MMA) in chloroform drops at pH 8.5 (AVI) Selective dissolution of p(BMA-co-DAMA-co-MMA)-rich shell from core-shell microcarriers at pH 5 (AVI) Selective dissolution of p(BMA-co-DAMA-co-MMA)-rich nut from acorn-shaped microcarriers at pH 5 (AVI) Negligible dissolution of inverted core-shell microcarriers at pH 5 (AVI) Sequential dissolution of p(BMA-co-DAMA-co-MMA)rich outer shell and core from triple microcarriers at pH 5 (AVI) Dissolution of p(BMA-co-DAMA-co-MMA)-rich matrix from microcarriers with structure of multicores-in-matrix 1437

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438

Article

Chemistry of Materials (14) Choi, C.-H.; Weitz, D. A.; Lee, C.-S. One Step Formation of Controllable Complex Emulsions: From Functional Particles to Simultaneous Encapsulation of Hydrophilic and Hydrophobic Agents into Desired Position. Adv. Mater. 2013, 25, 2536−2541. (15) Choi, D. H.; Park, C. H.; Kim, I. H.; Chun, H. J.; Park, K.; Han, D. K. Fabrication of core-shell microcapsules using PLGA and alginate for dual growth factor delivery system. J. Controlled Release 2010, 147, 193−201. (16) Kim, S.-H.; Shum, H. C.; Kim, J. W.; Cho, J.-C.; Weitz, D. A. Multiple Polymersomes for Programmed Release of Multiple Components. J. Am. Chem. Soc. 2011, 133, 15165−15171. (17) Shum, H. C.; Zhao, Y.; Kim, S.-H.; Weitz, D. A. Multicompartment Polymersomes from Double Emulsions. Angew. Chem. Int. Ed. 2011, 50, 1648−1651. (18) Trongsatitkul, T.; Budhlall, B. M. Multicore-Shell PNIPAm-coPEGMa Microcapsules for Cell Encapsulation. Langmuir 2011, 27, 13468−13480. (19) Zhang, H.; Liu, D.; Shahbazi, M.-A.; Mäkilä, E.; Herranz-Blanco, B.; Salonen, J.; Hirvonen, J.; Santos, H. A. Fabrication of a Multifunctional Nano-in-micro Drug Delivery Platform by Microfluidic Templated Encapsulation of Porous Silicon in Polymer Matrix. Adv. Mater. 2014, 26, 4497−4503. (20) Haase, M. F.; Brujic, J. Tailoring of High-Order Multiple Emulsions by the Liquid-Liquid Phase Separation of Ternary Mixtures. Angew. Chem. Int. Ed. 2014, 126, 11987−11991. (21) Zarzar, L. D.; Sresht, V.; Sletten, E. M.; Kalow, J. A.; Blankschtein, D.; Swager, T. M. Dynamically reconfigurable complex emulsions via tunable interfacial tensions. Nature 2015, 518, 520−524. (22) Hwang, D. K.; Oakey, J.; Toner, M.; Arthur, J. A.; Anseth, K. S.; Lee, S.; Zeiger, A.; Van Vliet, K. J.; Doyle, P. S. Stop-Flow Lithography for the Production of Shape-Evolving Degradable Microgel Particles. J. Am. Chem. Soc. 2009, 131, 4499−4504. (23) Wang, J.-Y.; Wang, Y.; Sheiko, S. S.; Betts, D. E.; DeSimone, J. M. Tuning Multiphase Amphiphilic Rods to Direct Self-Assembly. J. Am. Chem. Soc. 2012, 134, 5801−5806. (24) Min, N. G.; Kim, B.; Lee, T. Y.; Kim, D.; Lee, D. C.; Kim, S.-H. Anisotropic Microparticles Created by Phase Separation of Polymer Blends Confined in Monodisperse Emulsion Drops. Langmuir 2015, 31, 937−943. (25) Saito, N.; Kagari, Y.; Okubo, M. Effect of Colloidal Stabilizer on the Shape of Polystyrene/Poly(methyl methacrylate) Composite Particles Prepared in Aqueous Medium by the Solvent Evaporation Method. Langmuir 2006, 22, 9397−9402. (26) Wang, D.; Huang, X.; Wang, Y. Managing the Phase Separation in Double Emulsion by Tuning Amphiphilicity via a Supramolecular Route. Langmuir 2014, 30, 14460−14468. (27) Jeong, J.; Um, E.; Park, J.-K.; Kim, M. W. One-step preparation of magnetic Janus particles using controlled phase separation of polymer blends and nanoparticles. RSC Adv. 2013, 3, 11801−11806. (28) Loxley, A.; Vincent, B. Preparation of Poly(methyl-methacrylate) Microcapsules with Liquid Cores. J. Colloid Interface Sci. 1998, 208, 49−62. (29) Kim, S. W.; Hwangbo, K.-H.; Lee, J. H.; Cho, K. Y. Microfluidic fabrication of microparticles with multiple structures from a biodegradable polymer blend. RSC Adv. 2014, 4, 46536−46540. (30) Li, W.; Dong, H.; Tang, G.; Ma, T.; Cao, X. Controllable microfluidic fabrication of Janus and microcapsule particles for drug delivery applications. RSC Adv. 2015, 5, 23181−23188. (31) Ekanem, E. E.; Nabavi, S. A.; Vladisavljević, G. T.; Gu, S. Structured Biodegradable Polymeric Microparticles for Drug Delivery Produced Using Flow Focusing Glass Microfluidic Devices. ACS Appl. Mater. Interfaces 2015, 7, 23132−23143. (32) Belfiore, L. A. Physical Properties of Macromolecules; John Wiley & Sons, Inc.: Hoboken, NJ, 2010. (33) de Gennes, P. G. Dynamics of fluctuations and spinodal decomposition in polymer blends. J. Chem. Phys. 1980, 72, 4756− 4763.

(34) Kim, S.-H.; Abbaspourrad, A.; Weitz, D. A. Amphiphilic Crescent-Moon-Shaped Microparticles Formed by Selective Adsorption of Colloids. J. Am. Chem. Soc. 2011, 133, 5516−5524. (35) Nisisako, T.; Hatsuzawa, T. A microfluidic cross-flowing emulsion generator for producing biphasic droplets and anisotropically shaped polymer particles. Microfluid. Nanofluid. 2010, 9, 427−437. (36) Prasad, N.; Perumal, J.; Choi, C.-H.; Lee, C.-S.; Kim, D.-P. Generation of Monodisperse Inorganic−Organic Janus Microspheres in a Microfluidic Device. Adv. Funct. Mater. 2009, 19, 1656−1662. (37) Pannacci, N.; Bruus, H.; Bartolo, D.; Etchart, I.; Lockhart, T.; Hennequin, Y.; Willaime, H.; Tabeling, P. Equilibrium and Nonequilibrium States in Microfluidic Double Emulsions. Phys. Rev. Lett. 2008, 101, 164502. (38) Pekarek, K. J.; Jacob, J. S.; Mathiowitz, E. Double-walled polymer microspheres for controlled drug release. Nature 1994, 367, 258−260. (39) Torza, S.; Mason, S. G. Three-Phase Interactions in Shear and Electrical Fields. J. Colloid Interface Sci. 1970, 33, 67−83. (40) Hwang, S.; Lahann, J. Differentially Degradable Janus Particles for Controlled Release Applications. Macromol. Rapid Commun. 2012, 33, 1178−1183. (41) Uhrich, K. E.; Cannizzaro, S. M.; Langer, R. S.; Shakesheff, K. M. Polymeric Systems for Controlled Drug Release. Chem. Rev. 1999, 99, 3181−3198. (42) Warburg, O.; Posener, K.; Negelein, E. Ü ber den Stoffwechsel der Carcinomzelle. Biochem. Z. 1924, 152, 319−344. (43) Swain, J. E. Optimizing the culture environment in the IVF laboratory: impact of pH and buffer capacity on gamete and embryo quality. Reprod. BioMed. Online 2010, 21, 6−16. (44) Xiao, W.; Zeng, X.; Lin, H.; Han, K.; Jia, H.-Z.; Zhang, X.-Z. Dual stimuli-responsive multi-drug delivery system for the individually controlled release of anti-cancer drugs. Chem. Commun. 2015, 51, 1475−1478.

1438

DOI: 10.1021/acs.chemmater.5b04798 Chem. Mater. 2016, 28, 1430−1438