Reactive Microfluidic Production of Degradable Microgels for Drug Delivery

INTRODUCTION: Microgels, best defined as hydrogel particles, have great potential to address many challenges in biomedicine. The facile injectability, high surface area to volume ratio, and ready tunability of both the surface and bulk of microgels present key advantages for a range of applications including drug delivery or drug scavenging vehicles¹, cell encapsulation matrices², or biosensors³. Microgels prepared using “smart” materials that can dynamically change their physical properties (e.g. size, charge, or colloidal stability) in response to external factors (e.g. temperature, pH, or the concentration of a particular chemical in the microgel environment) have particular utility in this regard, facilitating “on-demand” changes in microgel localization (i.e. aggregation versus circulation) or drug diffusion in different environments⁴. Of particular note, microgel size has a significant impact on the ultimate biological performance of the microgel; for example, smaller microgels release drugs faster and circulate for an extended period of time (ultimately localizing in the spleen or liver) while larger microgels release drug slower and typically remain at their injection site. However, conventional methods of fabricating environmentally-responsive microgels are significantly limited in terms of their capacity to control microgel size. While monodisperse thermoresponsive microgels with sizes from 100-1000 nm are relatively easy to fabricate via a precipitation-based technique⁴, inverse emulsion strategies are typically required to synthesize non-thermoresponsive microgels or any microgel with a diameter of >1 micron. Conventional methods to generate inverse emulsions rely on sonication, homogenization, or other types of bulk shear forces to generate the water-in-oil droplets, resulting in polydisperse microgels that give unpredictable drug release kinetics and biological responses. However, microgels with >1 micron diameters are of significant technological interest in that they are less likely to be endocytosed by macrophages or quickly sequestered within the body upon injection. Of particular utility for local drug delivery would be monodisperse populations of large (>1 micron diameter) microgels crosslinked via degradable linkages; such materials could be injected at a desired site of action, remain at the site of injection to deliver drug locally over an extended period of time, and then cleared from the body after use.

EXPERIMENTAL: Microfluidics chips were designed that facilitated on-chip crosslinking between two reactive polymer precursors, pre-mixed prior to the droplet formation step. Microgels were generated by reacting hydrazide-functionalized polymers (prepared via EDC-mediated grafting of adipic acid dihydrazide with a carboxylic acid-functionalized polymer) with aldehyde-functionalized polymers (prepared via carbohydrate oxidation using sodium periodate) to spontaneously form a hydrazone-crosslinked microgel inside the microfluidics chip. The chip (pictured) contacts the two aqueous reactive polymer streams, passes the streams through a mixing channel, and then generates water-in-oil droplets using paraffin oil as the oil phase and Span 80 as the surfactant. Microgels were then extracted into an aqueous suspension via freeze-thawing and subsequent solvent exchange. Particle size of the microgels was assessed via light and fluorescence microscopy and surface charge was measured via electrophoretic mobility measurements. Drug loading/release experiments were performed by mixing purified microgels in a drug solution, centrifuging to isolate the microgels, resuspending the microgels in PBS inside a Float-a-Lyzer membrane bag (Spectrum Labs, 50,000 MWCO), and assaying drug release as a function of time. Microgels with the same composition were prepared using sonication to compare with microgels synthesized using microfluidics.

RESULTS: Reactive microfluidics facilitates successfully production of microgels with significantly improved monodispersity and size control relative to traditional shear-based techniques. Microgels with particle sizes ranging between 25-200 microns were generated with polydispersities <5% by varying both the relative flow rates of the oil and (aqueous) polymer streams or the total flow rate of the oil and polymer streams through the chip. Crosslinked microgels were generated spontaneously upon droplet formation without the need for subsequent heating, UV photopolymerization, or the use of additional initiators, crosslinkers, or chain transfer agents, as required in previous microfluidic microgel generation systems⁵. Continuous particle production was possible over at least 72 hours, with increased microgel yield possible through the use of parallel microfluidic reactors. Microgels were successfully synthesized based on carbohydrates (carboxymethyl cellulose/dextran), mixtures of synthetic oligomers and carbohydrates (poly(N-isopropylacrylamide)/ dextran) or synthetic oligomers only (poly(N-isopropylacrylamide)), illustrating the adaptability of the device to generate multiple types of degradable microgels with varying chemistries. Drug release for the model drug bupivacaine (a local anesthetic) was a weak function of microgel size and followed first-order, diffusion-based kinetics, with sustained release over periods of up to two weeks.

CONCLUSIONS: Reactive microfluidics provides an effective route to the preparation of monodisperse degradable microgels based on natural, synthetic, or mixtures of natural and synthetic polymers. Such materials are ideal for biomedical applications in that the crosslink density, composition, size, and biocompatibility can be tuned appropriately for the particular desired end-use.

ACKNOWLEDGEMENTS: This work was funded by the Natural Sciences and Engineering Research Council of Canada and 20/20: NSERC Ophthalmic Materials Research Network.

REFERENCES: (1) (a) Eichenbaum, G. M.; Kiser, P. F.; Dobrynin, A. V.; Simon, S. A.; Needham, D. Macromolecules 1999, 32, 4867(b) Hoare, T.; Pelton, R. Langmuir 2008, 24, 1005; (2) Bae, K. H.; Yoon, J. J.; Park, T. G. Biotechnology Progress 2006, 22, 297; (3) Das, M.; Zhang, H.; Kumacheva, E. Annual Review of Materials Research 2006, 36, 117; (4) Hoare, T.; Pelton, R. Macromolecules 2004, 37, 2544; (5) (a) Kim, J. W.; Utada, A. S.; Fernandez-Nieves, A.; Hu, Z. B.; Weitz, D. A. Angewandte Chemie-International Edition 2007, 46, 1819(b) Zhang, H.; Tumarkin, E.; Sullan, R. M. A.; Walker, G. C.; Kumacheva, E. Macromolecular Rapid Communications 2007, 28, 527.