Natural polymers include proteins (silk, collagen, etc.), polysaccharides (alginate, chitosan, cellulose, etc.), and polynucleotides (DNA and RNA). Especially, two natural polysaccharides: alginate and chitosan, are widely utilized in biomedical and bioengineering fields.

Multiple synthetic polymers have been used to construct nano- or microgels. Especially, we discussed two synthetic polymers: poly(N-isopropylacrylamide) and poly (ethylene glycol), due to their stimulus responsiveness and bio-compatibility.

Emulsion polymerization generally involves an aqueous phase containing surfactants, an oil phase containing monomers and crosslinkers, and initiators to start the radical polymerization process (Figure 1A; Tobita et al., 2000). The system is then homogenized to generate droplets of monomer in oil phase, surrounded by the water phase. The surfactants stabilize the droplets and prevent them from aggregation. By initiating the reaction (e.g., heating the system for a thermal initiator), the initiators react to form free radicals that start the polymerization process. In a variant of this technique (inverse emulsion polymerization), the aqueous phase contains the monomers and crosslinkers while the initiators can be in either of two phases. The particle size can be tuned by adjusting parameters such as the homogenization speed and the reaction temperature (Tobita et al., 2000). Biologics, such as functional proteins or drugs can be loaded by incubation with collected nano- or microgels after synthesis (Liu and Garcia, 2016). The system for emulsion polymerization may also be adjusted to remove the heating step, or the surfactants, to facilitate the biologics loading.

Resembling emulsion polymerization, precipitation polymerization also proceeds in a batch process. However, precipitation polymerization starts in a continuous phase, where the monomers and initiators are completely soluble, and no stabilizing agents are needed (Figure 1B). After initiating the reaction (e.g., increasing the system temperature), a spontaneous nucleation process occurs and polymerization proceeds.

Lithographic technique fabricates nano- or microgels by templating hydrogels at nano- or micro-scale level. This method needs a template to control both the size and morphology of the product (Helgeson et al., 2011). For example, imprint lithography utilizes a template that acts as a mold for the hydrogel precursors, and UV light is then applied to polymerize the precursors inside the mold (Figure 1C). Nano- or microgels are recovered from the mold, typically by mechanical delamination post-synthesis. Biologics are often loaded by incubation with collected nano- or microgels after polymerization to avoid the UV light exposure. Alternatively, biologics can also be mixed with the hydrogel precursors if the crosslinking condition is mild. Lithographic technique provides a precise control over particle size, morphology, and monodispersity by tuning the template characteristics.

Microfluidic synthesis technique requires a lithographically fabricated microfluidic device (typically manufactured by polydimethylsiloxane), and generates nano-, or microgels droplets one at a time in a continuous manner. Taking emulsion-based microfluidic system as an example, multiple phases (monomers in aqueous, crosslinkers in oil, etc.) meet in a junction geometry and droplets form. Crosslinking will occur right after the formation of these droplets (Figure 1D). The biologics are often loaded by mixing with the monomers in the aqueous phase. Characteristics of nano- or microgels, including size, morphology, and size distribution can be precisely controlled by several parameters, such as the nozzle diameter and flow rate (Headen et al., 2014).

In the electrospray, the liquid flows from a capillary nozzle through an electric field that disrupts a large droplet into nano- or microdroplets. The nano- or microdroplets are then collected and homogenized in the crosslinking solution (Figure 1E). When an electric field is applied to a droplet, the electric charge generates an electrostatic force into the droplet. Nano- or microdroplets will form when the electrostatic force overcomes the cohesive force of the droplet. The particle size and the size distribution can be tuned by adjusting parameters, such as the flow rate of the solution and the applied electric potential (Tapia-Hernandez et al., 2015). Electrospray has been developed into a mature technology with encapsulation systems (e.g., alginate/Ca²⁺), and is particularly convenient to encapsulate medicines, foods, and microorganisms.

Besides the general requirements for a drug delivery system, such as optimum loading capacity and biocompatibility, effective delivery of drugs at correct time and location is always considered a big challenge. It requires the system to sense the signals (environmental sensitivity) and actively respond to them (controlled release). Nano- or microgels are potential candidates since the materials can be engineered to sense the environmental cues, while the release of the drugs can be triggered by the volume change of materials (Figure 2A), or the pH induced hydrolysis (Liu et al., 2014; Yu et al., 2016).

Responsive nano or micro-drug delivery systems have been extensively studied in cancer therapy. Due to the weak acidic micro-environment of tumors, pH-responsive nano, or microgels carriers with acid-cleavable bond are usually designed to ensure the controlled drug release. Xiong et al. reported a drug carrier, poly(N-isopropylacrylamide-co-acrylic acid) nanogel with dual temperature and pH sensitivity (Xiong et al., 2011). The system precipitated on the heated cancer tissues (~42°C, local hyperthermia) due to the hydrophilic to hydrophobic transition when temperature was higher than its LCST. The anti-cancer drug doxorubicin (DOX) was covalent linked to the nanogel via an acid-cleavable bond, which could be released efficiently in the acidic microenvironment of tumor tissues. The system utilized both the temperature and pH sensitivity of the materials, as well as acid-cleavable design to ensure the controlled release missions.

In another work, Song et al. developed a responsive nanogel system to deliver both a chemotherapeutic drug (paclitaxel) and a cytokine (interleukin-2) to combat tumors (Song et al., 2017). They first modified the chitosan into two oppositely charged chitosan derivatives. The nanogels were then assembled by mixing these two chitosan derivatives, and further photo-crosslinked with addition of hydroxypropyl-β-cyclodextrin acrylate. The pH responsiveness of nanogel to a weak acidic tumor microenvironment could be fine-tuned by adjusting the formulation, while incorporation of hydroxypropyl-β-cyclodextrin acrylate increased the encapsulation efficiency of paclitaxel (very low solubility in aqueous phase). This responsive bio-hybrid nanogel system integrated chemotherapy and immunotherapy, and significantly enhanced the antitumor activity with improved calreticulin exposure and antitumor immunity.

These responsive nano- or microscale biohybrid system are also widely used in controlled diabetic therapeutic delivery. For example, Sung et al. developed a pH-responsive system consisting of chitosan and poly(γ-glutamic acid) for oral insulin delivery (Sung et al., 2012). Poly(γ-glutamic acid) was incorporated since it could conjugate with insulin via Zn²⁺ to enhance the loading efficiency. The acidic environment of gastric medium in the stomach stabilized the system due to the ionization of chitosan backbones. At intestine (pH ~ 7.0–8.0), the chitosan chains were deprotonated and collapsed. The swollen to shrunk transition of the encapsulating materials enabled the insulin release.

The stimulus-responsive features of nano- or microgels can be further coupled with other biologics, such as an enzyme, to achieve a more sophisticated signal-sensing capability. Gu et al. reported an injectable microgels for controlled glucose-responsive release of insulin by integrating a pH-responsive chitosan matrix, enzyme nanocapsules and insulins (Gu et al., 2013). The enzyme converted the glucose into the gluconic acid at the hyperglycemic condition, and therefore swelled the chitosan matrix due to the protonation of the chitosan networks. Consequently, these microgels were self-regulating and able to release insulins at the basal or higher rate based on normal or hyperglycemic conditions.

Not only materials, engineered cells can also respond to the external signals accordingly (Figure 2B). Indeed, cell-based therapies are considered as one of the most promising next generation medicines. Various synthetic gene circuits have been assembled to treat diverse diseases, including metabolic disorders, cancer and immune diseases (Bulter et al., 2004; Higashikuni et al., 2017; Xue et al., 2017).

Wang et al. reported a pain management strategy based on microencapsulated cell-engineering principles (Wang et al., 2018). The engineered cells can respond to volatile spearmint aroma and produce an analgesic peptide (huwentoxin-IV) that selectively inhibits the pain-triggering voltage-gated sodium channel. Engineered cells were encapsulated in alginate capsules (400 μm). Their results showed that mice (chronic inflammatory and neuropathic pain model) implanted with the capsules demonstrated reduced pain-associated behaviors on oral or inhalation-based intake of spearmint essential oil, with negligible cardiovascular, immunogenic, and behavioral side effects.

In another example, Shao et al. demonstrated a smartphone-assisted treatment of diabetes in mice by microgel encapsulated cells (Shao et al., 2017). In their design, the implanted capsules carried both optogenetically engineered cells and wirelessly powered far-red light (FRL) LEDs (light-emitting diodes). The far-red light LEDs were remotely controlled by smartphone programs or bluetooth-active glucometer in a glucose-dependent manner. Optogenetically engineered cells could respond to FRL based on the bacterial light-activated cyclic diguanylate monophosphate (c-di-GMP) synthase, and activate the expression of mouse insulin. The mouse insulin would then diffuse out from the capsules to control the glucose level in blood.

Quorum sensing (QS) refers to the ability of organisms to detect and respond to the population density with a specific behavior (Figure 2C; Scott et al., 2017; Shuma and Balazs, 2017). QS plays an essential role in the life cycle of bacteria, yeast, as well as social insects. Designing a man-made system mimicking QS behavior, that is sensing and responding to the system size and density is important but challenging (Ford and Silver, 2015; Li et al., 2017; Shuma and Balazs, 2017). In this notion, polymeric microgels can be integrated with QS cells, since the microgels provide a natural spatial segregation in differentiating the interior and exterior environments. Therefore, the cell populations in individual microgel are effectively insulated from the surrounding environment and other microgels. The porous structure traps the cells, but allows the free diffusion of nutrients and signaling molecules.

In this notion, Huang and Lee demonstrated an engineering safeguard to prevent unintended proliferation by coupling collective survival and environment sensing of bacteria (Lopatkin et al., 2016). Programmed by an engineered circuit, the cells can produce beta-lactamase (BlaM), which is able to degrade beta-lactam antibiotics, such as carbenicillin (Cb) only at a sufficiently high density regulated by QS. At the same time, sufficient cells were required to produce and secrete enough BlaM for population survival. Consequently, the cells inside the microgel can sense the physical confinement, produce BlaM and get rescued due to their high densities. Those escaping from microgel, instead, will be eliminated.

Nano- or microgels have been used as building blocks for scaffolds due to its several advantages including ease of fabrication and rapid response to stimulus (Gan et al., 2009; Dhandayuthapani et al., 2011). Cell adhesion on two-dimensional surface or three-dimensional scaffold has been a focus of biophysical research (Dai and Ngai, 2013). Multiple physical or chemical features of the nano- or microgels can dictate their interactions with cells, including the composition, physical properties (porosity, stiffness, etc.) and topography (Leach and Whitehead, 2018). These interactions will direct cells morphogenesis, proliferation, or even differentiation.

Mesenchymal stem cells (MSCs) have been a focus in cell-based therapies for tissue repair and regeneration because of their multilineage differentiation potentials (Augello et al., 2010; Boeuf and Richter, 2010). Growth factors and other chemical inductive cues can effectively induce MSCs differentiation. However, this approach suffers from its own restrictions including the potential off-target effects when treated in large dosages, and long-term maintenance of phenotype after the removal of these cues (Leach and Whitehead, 2018). It has been proven that extracellular matrix (ECM) can direct the MSCs fate through the physical interactions (Engler et al., 2006). Multiple studies have used nano- or microgels as building blocks to assemble the extracellular matrix (ECM) and investigate their effects on the MSCs differentiation (Figure 2D).

From composition perspective, Dai et al. demonstrated that the interactions between MSCs and pNIPAM-AA (poly-N-isopropylacrylamide-acrylic acid) microgels can have a direct effect on osteogenesis of MSCs. Due to the existence of carboxyl groups in the microgels, supplementing microgels could either absorb the free calcium ions and prevent the osteogenesis (supplementing during early osteogenesis), or bind the calcium deposited cells (supplementing during the late osteogenesis) to further promote the osteogenesis through its interactions with cells.

In another work, Caldwell et al. assembled the PEG (polyethylene glycol) microgels into a microporous, covalently linked material, and seeded the human mesenchymal stem cells (hMSCs) into the porous scaffold (Caldwell et al., 2017). To assemble the scaffold, two PEG microgels functionalized with DBCO (dibenzocylcooctyne) or azide groups were separately prepared, and then mixed for crosslinking. Especially, they used two conditions (low shear and high shear) to generate microgels with different sizes. Microgels formed using low shear (votex) had a mean particle diameter of 120 μm (PEG microgels with DBCO) and 130 μm (PEG microgels with azide). Microgels formed using high shear (sonication) were an order of magnitude smaller, with average particle sizes of 16 μm (PEG microgels with DBCO), and 15 μm (PEG microgels with azide). While cells showed high survival rate in both cases, their morphology differed significantly in two scaffolds made by small or large microgels. The cells seeded in the networks made of small microgels had only small protrusions into the matrix, with only diffuse actin fibers present. Comparatively, in the networks made by the big microgels, hMSCs spread and exhibited visible actin fibers.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.