Single Versus Dual Microgel Species for Forming Guest-Host Microporous Annealed Particle PEG-MAL Hydrogel

Abstract

Inter-particle secondary crosslinks allow microporous annealed (MAP) hydrogels to be formed. Methods to introduce a secondary crosslinking network in MAP hydrogels include particle jamming, annealing with covalent bonds, and reversible non-covalent interactions. Here, we investigate the effect of two different approaches to secondary crosslinking of polyethylene glycol (PEG) microgels via reversible guest-host interactions. We generated a dual-particle MAP-PEG hydrogel using two species of PEG microgels, one functionalized with the guest molecule, adamantane, and the other with the host molecule, β-cyclodextrin. In a different approach, a mono-particle MAP-PEG hydrogel was generated using one species of microgel functionalized with both guest and host molecules. The dual-MAP-PEG formed small, clustered domains of microgels of each type, rather than a well-mixed distribution. The mono-MAP-PEG formed a homogenous distribution due to the single type of microgels used. We then compared the mechanical properties of these two types of MAP-PEG hydrogels and found that mono-MAP-PEG resulted in significantly softer gels with a lower yield stress. We speculate this is because functionalization with both guest and host molecules in mono-MAP-PEG enabled intra-particle guest-host bonding. Based on these studies, mono-MAP-PEG provides a homogeneous guest-host hydrogel that is soft and injectable for tissue engineering.

Introduction

Microporous annealed particle (MAP) hydrogels are a platform for engineering cellular microenvironments with intrinsic interstitial spaces (1–4). MAP hydrogels are composed of closely packed microgels to create porosity in the negative space between individual microgels. Researchers have demonstrated the utility of MAP hydrogels as excellent biomaterial environments for organoid culture, wound healing, tissue regeneration/vascularization, and rapid cell migration (5–7). Furthermore, MAP hydrogels are generally injectable, self-healing, and shear-thinning, as individual microgels slide past one another under shear and re-assemble when the shear force is discontinued (8). Microgels that makeup MAP hydrogel can be engineered with custom bioactive and material properties similar to larger bulk hydrogels (1, 9). Multiple types of microgels can be combined in a “building block” approach to generate mixed MAP hydrogels with multiparameter bioactivity, drug delivery, and mechanical properties (4).

We and others have demonstrated that microgel particles can be interlinked using secondary crosslinking mechanisms, stabilizing the overall MAP hydrogel structure (2, 7, 8, 10, 11). Examples of secondary crosslinking mechanisms include covalent interparticle annealing by methods such as photo-crosslinking, enzymatic crosslinking, and click chemistry (6, 11–13). However, permanent crosslinks limit the time window for injectability to pre-crosslinked gels and require a post-delivery annealing step. Alternatives to covalent annealing are reversible crosslinking mechanisms such as guest-host interactions, hydrogen bonding, ionic interactions, metal-ligand coordination, and dynamic hydrazone bonds (7, 8, 14–17).

Guest-host interactions are non-covalent, shear-reversible interactions that are composed of a guest and host molecule. One of the most well-characterized guest-host interactions is the β-cyclodextrin (β-CD) / adamantane (Ada) interaction (7, 10, 18–24). β-CD is a cyclic oligosaccharide of repeating D-glucose units that forms a cup-like morphology with a hydrophobic core. Adamantane is a small hydrophobic hydrocarbon that acts as the guest and can interact with the hydrophobic core of the β-cyclodextrin (24). Previously, we designed polyethylene glycol maleimide (PEG-MAL) MAP hydrogel interlinked with guest-host interactions (7, 10). The MAP hydrogel was formed by mixing two different populations of microgels, one conjugated with Ada and the other with β-CD. However, due to the design involving two different microgel species, we observed grain boundaries between gel populations, and microgel clusters in microdomains in the MAP hydrogel structure (7, 10).

In this study, we improved on the guest-host MAP-PEG hydrogel design by incorporating both guest and host molecules on a single PEG microgel species, thus eliminating the two microgel species separation problem. However, the incorporation of both guest and host molecules on a single microgel also introduced intra-particle guest-host reactions, the effect of which on the overall mechanical properties is difficult to predict. Here, we evaluated the mechanical and rheological properties of the single-species design and compared it to the previous two-species design.

Results

We investigated the effect of guest-host secondary crosslinking on mono and dual-species granular hydrogels. The mono species hydrogels have both intra-and-inter-particle guest-host interactions while dual-species hydrogels have only inter-particle guest-host interactions (Figure 1A). We investigated the guest molecule adamantane and the host molecule β-cyclodextrin. We chose PEG-MAL as a model hydrogel to make our MAP scaffolds. PEG-MAL is highly reactive due to the maleimide groups on the arms of the PEG polymer. These maleimide groups react with thiols in click reactions with good cross-linking efficiency and rapid reaction speeds (25).

Figure 1: Synthesis of Guest-Host MAP-PEG.

(A) Schematic showing Ada-β-CD guest-host interactions. Ada (pink) interacts with the hydrophobic core of the β-CD (cyan) cup. Under increasing shear, the guest-host interaction dissociates and self-heals when the shear force is removed. (B) Scheme of Inter-GH-MAP-PEG synthesis. PEG-4MAL macromer is first functionalized with 1-adamantane-thiol and mono-thiol-β-cyclodextrin through michael-type addition chemistry. The PEGylated Ada (top) and PEGylated β-CD (bottom) are then added to a crosslinker (PEG-4SH) and emulsified into microgels that are later mixed to create the heterogeneous guest-host microgel structure. (C) Scheme of Intra-MAP-PEG synthesis. PEG-4MAL macromer is functionalized with both 1-adamantane-thiol and mono-thiol-β-cyclodextrin and emulsified with PEG-4SH crosslinker to create a homogeneous guest-host microgel structure.

We generated two different types of MAP-PEG hydrogel with secondary crosslinking: (1) one with inter-particle only interactions between two microgel species each functionalized with either a guest or host molecule (Inter-MAP-PEG) and (2) one with both intra- and inter-particle guest-host interactions where all microgels are homogeneously functionalized with both guest and host molecules (Intra-MAP-PEG) (Figure 1B, C). Synthesis of guest-host precursor molecules was performed by conjugating 1-adamantane-thiol (Ada-SH) and mono-thiol-β-cyclodextrin (β-CD-SH) to the PEG-4MAL before crosslinking in accordance with our previously published methods (7, 10). For Inter-MAP-PEG, Ada-SH and β-CD-SH were added to separate batches of the PEG-4MAL and allowed to react for 30 minutes before crosslinking with PEG-4SH in a water-in-oil emulsion. These two species of microgels (Ada-PEG and β-CD-PEG) were then combined, vortexed, and compacted to create the inter-particle secondary crosslinking network (Figure 1B). The Intra-MAP-PEG was created by functionalizing a single batch of PEG-4MAL with both adamantane-thiol and mono-thiol-β-CD, followed by crosslinking with PEG-4SH in a water-in-oil emulsion (Ada-β-CD-PEG Microgels) (Figure 1C).

We compared the structure of the Inter-MAP-PEG, Intra-MAP-PEG and the MAP-PEG without any guest-host interactions by confocal microscopy. For Inter-MAP-PEG, Ada-PEG microgels were labeled with Alexa Fluor-MAL-488 and β-CD-PEG microgels were labeled with Alexa Fluor-MAL-568. Intra-MAP-PEG and MAP-PEG were split into two populations and labeled with Alexa Fluor-MAL-488 or Alexa Fluor-MAL-568. We then mixed the two populations of each gel to determine the structure of each granular hydrogel scaffold, to control the effect of mixing artifacts that might be found in the dual-species Inter-MAP-PEG. We found that when the Ada-PEG microgels and the β-CD-PEG microgels were mixed and compacted into a granular hydrogel scaffold, they created even distributions of Ada or β-CD PEG microgels without clustering into microdomains (Figure 2A, B). The sizes between the Ada-PEG microgels and the β-CD-PEG microgels were consistent with a mean diameter of around 14.5 μm and a median diameter of 11.5 μm (Figure 2C). Intra-MAP-PEG and the MAP-PEG also demonstrated even mixing between the two species of microgels, indicating that neither the presence of guest-host interactions on the microgels nor the mixing of two different species of microgels had an effect on the overall structure of the granular hydrogel (Figure 2D–H). The mean diameter of the Intra-MAP-PEG microgels was 14.9 μm, and had a median diameter of 12.5 μm (Figure 2F). The mean diameter of the MAP-PEG microgels with no guest-host functional groups was 9.6 μm, with a median diameter of 8.9 μm (Figure 2E). The difference between the MAP-PEG microgel size and the size of the Inter-MAP-PEG and Intra-MAP-PEG is thought to be due to batch-to-batch variability in batch emulsion methods.

Figure 2: Visualization of microgel structure.

(A) 3D rendering of Inter-MAP-PEG after packing, Ada-PEG microgels labeled with Alexa Fluor-468 are pictured in magenta and β-CD-PEG microgels labeled with Alexa Fluor-488 are pictured in cyan. The microgels demonstrate an even distribution of Ada-PEG microgels and β-CD-PEG microgels. (B) Inter-MAP-PEG microgels in a large volume, both the Ada-PEG-microgels and the β-CD microgels separate into individual microgels without clustering. Scale bar is 100 μm. (C) The distribution of Inter-MAP-PEG microgels based on Feret’s diameter. Ada-PEG microgels and β-CD microgels had a similar distribution with a mean microgel diameter of 14.5 μm and a median microgel diameter of 11.5 μm. (D) 3D rendering of Intra-MAP-PEG after packing. Intra-MAP-PEG was split into two populations and labeled with separate dyes to control the mixing of the Inter-MAP-PEG hydrogels. The microgels demonstrate an even distribution between the Ada-β-CD-PEG-488 and the Ada-β-CD-PEG-568. (E) Intra-MAP-PEG microgels in a large volume. The scale bar is 100 μm. (F) The size distribution of Intra-MAP-PEG microgels. There was no significant difference between the size of the Ada-β-CD-PEG-488 microgels and the Ada-β-CD-PEG-568 microgels. The mean microgel diameter of both populations was 14.9 μm, and the median diameter was 12.5 μm. (G) 3D Rendering of MAP-PEG after packing with an even distribution of PEG-588 and PEG-568 microgels. (H) MAP-PEG microgels dispersed in a large volume. The scale bar is 100 μm. (I) The size distribution of MAP-PEG microgels. There was no significant difference between the PEG-488 and PEG-568 microgels. The mean microgel diameter of MAP-PEG microgels was 9.6 μm, and the median diameter was 8.9 μm.

We confirmed the presence of guest-host interactions on the Inter-MAP-PEG and Intra-MAP-PEG by incubation with soluble Alexa Fluor 647-maleimide conjugated to mono-thiol-β-cyclodextrin (β-CD-AF647). We found that Inter-MAP-PEG and Intra-MAP-PEG bound significantly more β-CD-AF647 than unfunctionalized microgels (Figure 3A), indicating that guest-host interactions were available to react on the surface of both the Inter and Intra-MAP-PEG scaffolds. We then analyzed the colocalization of the β-CD-AF647 with Ada-PEG microgels and β-CD-PEG microgels in the Inter-MAP-PEG scaffold, the Intra-MAP-PEG scaffold, and the MAP-PEG scaffold. (Figure 3B). We found that the β-CD-AF647 was more colocalized to Ada-PEG microgels (Pearson’s Coefficient = 0.43) than β-CD-PEG microgels (Pearson’s Coefficient = −0.32). The Intra-MAP-PEG microgels bound significantly less β-CD-AF647 than the Inter-MAP-PEG and both the Inter and Intra-MAP-PEG bound significantly more than the unfunctionalized MAP-PEG. The reverse experiment to determine the binding of Ada-Alexa Fluor to the microgels was not feasible as monomeric adamantane has poor water solubility. We also quantified the Mean Fluorescence Intensity (MFI) and found significantly higher β-CD-AF647 bound to the guest-host functionalized Inter- and Intra-MAP-PEG than the unfunctionalized base MAP-PEG (Figure 3C). Together, these data are consistent with β-CD-AF647 interacting with available Adamantane molecules and indicate that availability of guest-host bonding pairs on microgel surfaces.

Figure 3: Confirmation of guest-host interactions through colocalization.

(A) Colocalization of Inter-MAP-PEG microgels (β-CD-PEG microgels = cyan, Ada-PEG-microgels = magenta), Intra-MAP-PEG microgels (magenta), and unfunctionalized MAP-PEG microgels (magenta) with β-CD-AF647 (green). (B) Pearson’s coefficient of colocalization for Ada-PEG microgels and β-CD-PEG microgels in the Inter-MAP-PEG hydrogel, Intra-MAP-PEG, and MAP-PEG with β-CD-AF647. (C) Normalized mean fluorescence intensity for β-CD-Alexa Fluor 647 with Inter, Intra, or unfunctionalized MAP-PEG. Data are displayed as mean ± s.e.m. Significance calculated by one-way ANOVA, * = p < 0.05, ** = p < 0.01.

We investigated the swelling ratio of the hydrogels to identify differences in network crosslinking. We calculated the swollen polymer volume fraction for Inter-MAP-PEG, Intra-MAP-PEG, and unfunctionalized MAP-PEG after swelling for 24 hours. We found that there were no significant differences between the groups, and that they all behaved with low polymer volume fractions (Figure 4A) (26). These data indicate that functionalized with Ada and β-CD did not compromise the bulk hydrogel network structure.

Figure 4.

(A) Swelling ratio of swollen polymer content over initial polymer content for bulk hydrogel of the same composition as β-CD-PEG, Intra-MAP-PEG, and unfunctionalized MAP-PEG. (B) Determination of guest-host interactions on bulk hydrogels. Panel 1: Before contact, panel 2: after contact for Ada-β-CD-PEG bulk hydrogels, Ada-PEG and β-CD-PEG bulk hydrogels, and β-CD-PEG to β-CD-PEG bulk hydrogels as a control. (C) Extrusion of Intra-MAP-PEG through a 20 G needle to measure force of injection. (D) The extrusion force of Inter-MAP-PEG, Intra-MAP-PEG, and MAP-PEG. The maximum force that the force sensor could measure is indicated by the dashed line. (E) Wall shear stress calculated from the derived Navier Stokes equations.

Next, we manually tested for guest-hosting bonding between two bulk hydrogels. Functionalized PEG-4MAL macromers were cast into 6 mm silicone isolators and crosslinked by pipetting up and down with PEG-4SH. Bulk hydrogels with the same composition as the Intra-MAP-PEG were tested for the presence of guest-host interactions by bringing the gels into contact and testing for adhesion. We found that when the two Intra-MAP-PEG hydrogels were brought into contact, the gels adhered strongly enough to resist gravity when one half was lifted with forceps (Figure 4B). Inter-MAP PEG had the same response between Ada-PEG hydrogel and β-CD-PEG hydrogel, but two hydrogels functionalized with the same molecule (e.g., β-CD-PEG to β-CD-PEG) had no adhesion (Figure 4B). These data further support the availability and specificity of guest-host interactions for inter-microgel secondary crosslinking for both Inter-MAP-PEG and Intra-MAP-PEG.

An advantage of granular hydrogels is their injectability. We tested the injection forces and wall shear stresses of the material by extruding the MAP hydrogels through a 20 G needle and measuring the force of injection (Figure 4C). We found that there were no significant differences for the maximum extrusion force for the Inter-, Intra-, or MAP-PEG (Figure 4D). We then used the force of injection data and needle geometry to derive the wall shear stress within the needle during injection for each material. There were no significant differences between the wall shear stress between the materials, and each had similar shear stress to physiological shear stress found within microvasculature (~ 9.5 Pa) (Figure 4E) (27). We found that all materials underwent plug flow within the center of the needle, with laminar flow at the edges of the needle, representative of a turbulent flow for a Herschel-Bulkley fluid (28, 29).

To quantify the effect of inter- and intra-particle guest-host interactions on PEG microgel mechanical properties, we performed a series of rheological tests. Inter-MAP-PEG was compared to Intra-MAP-PEG and unfunctionalized MAP-PEG. We performed a low amplitude (1%) oscillatory frequency sweep from 10 to 0.01 Hz. We found that all groups underwent viscoelastic behavior over the entire frequency range (Figure 5A) determined by a greater storage modulus (G’) than loss modulus (G”). Inter-MAP-PEG and MAP-PEG had a significantly higher storage modulus (Inter-MAP-PEG G’: 1330 Pa, MAP-PEG G’: 1330 Pa) than Intra-MAP-PEG (G’: 530 Pa) over the frequencies observed. We next conducted an oscillatory strain sweep from 0.001 % to 500 % at a constant frequency of 1 Hz and calculated the approximate yield shear stress for Inter-MAP-PEG (470 Pa), Intra-MAP-PEG (90 Pa), and MAP-PEG (470 Pa) (Figure 5B). Next, we performed a cyclic strain sweep alternating between low (1%) and high (500%) strains to understand the shear-thinning and recovery characteristics of the different microgel groups. We found that all groups could recover after a period of high strain (Figure 5C). To further investigate how guest-host interactions influence the yielding behavior we conducted a unidirectional shear rate sweep. All materials displayed shear-thinning behavior as shear rate increased. We determined the yield stress of the different groups using a linear regression. We found that the yield stress of Inter-MAP-PEG (80 Pa) and MAP-PEG (100 Pa) was significantly higher than Intra-MAP-PEG (40 Pa) (Figure 5D, E). Overall, the Intra-MAP-PEG was significantly softer than the Inter-MAP-PEG or unfunctionalized MAP-PEG. We also found that the yield stress, or the stress required to initiate flow in a material, for Intra-MAP-PEG was significantly lower than that of the Inter-MAP-PEG and the unfunctionalized MAP-PEG.

Figure 5:

(A) Frequency sweeps from 10 to 0.01 Hz at 1 % amplitude for Inter-MAP-PEG (blue), Intra-MAP-PEG (pink), and unfunctionalized MAP-PEG (orange). All gels demonstrate viscoelastic behavior over the frequency range. (B) Oscillatory strain sweep from 100 % to 0.001 %. (C) Cyclic strain alternating between low (1 %) for 120 s and high (500 %) strain for 60 s. (D) Unidirectional shear rate sweeps from 100 1/s to 0.01 1/s. (E) Yield stresses are calculated from a linear regression of the unidirectional shear rate sweeps.

We then investigated the effect of polymer weight percentage (wt %) on the mechanical properties of Intra-MAP-PEG and Inter-MAP-PEG. We measured yield stress by unidirectional flow shear rate rheology. We found that all groups had increasing yield stress with increasing wt% of the polymer. (Figure 6A–C). With a guest-host concentration of [0.25 mM] of Ada and [0.25 mM] β-CD in the Intra-MAP-PEG the mechanical properties of all wt % samples were lower than the Inter-MAP-PEG and the Unfunctionalized MAP-PEG. Since the change in the polymer concentration was not significant in the mechanical properties of the Intra-MAP-PEG compared to the Inter-MAP-PEG and MAP-PEG groups, we determined that the effect of the concentration of guest-host interactions played a greater contribution.

Figure 6: Rheology of different wt % microgels.

Panel 1: Unidirectional shear rate sweep from 100 1/s to 0.01 1/s for 4, 6, 8, and 10 wt % microgels. Panel 2: Calculated yield stress from unidirectional shear rate sweep for (A) Inter-MAP-PEG, (B) Intra-MAP-PEG, 0.25 mM concentration of Ada and β-CD, and (C) MAP-PEG. Significance between means or three or more groups determined by one-way ANOVA with Tukey’s post-hoc pairwise comparisons. * = p < 0.05, ** = p < 0.01, *** = p < 0.001, **** = p < 0.0001.

To ensure that the differences in shear thinning and flow behavior between the microgels was not due to swelling in different weight percentages, we quantified the sizes of the swollen microgels (Supplementary Figure 1). We found that there was no significant difference between the swollen microgels based on weight percentage. We continued to study the bulk hydrogel equivalents of these groups and altered the wt % to determine if there were any significant differences between the microgels and the bulk hydrogels. We studied the shear-thinning behavior of β-CD-PEG bulk hydrogel, Intra-MAP-PEG (Ada-β-CD-PEG) bulk hydrogel, and unfunctionalized MAP-PEG (PEG-MAL) bulk hydrogels through a unidirectional flow shear rate rheology (Supplementary Figure 2A, B, C). Much like with the microgels, we found that all bulk hydrogels had a higher yield stress with increasing weight percentage. We found that 6, 8, and 10 wt% Intra-MAP-PEG bulk hydrogels had a greater yield stress than both the β-CD-PEG hydrogel and the unfunctionalized MAP-PEG bulk hydrogel. This suggests a contribution of the guest-host interactions to the mechanical strength within the bulk of the hydrogel structure.

We had originally hypothesized that due to the more homogeneous structure of the inter-particle guest-host interactions in Intra-MAP-PEG, there would be more guest-host interactions throughout the scaffold as compared to the Inter-MAP-PEG. We titrated the concentration of guest-host functionalization on the PEG-MAL macromer to determine the effect of guest-host particles as an intra-particle secondary crosslinking network. We conducted a low amplitude (1%) oscillatory frequency sweep from 0.1 to 10 Hz for the different concentrations. The concentrations ranged from 0.5 mM (0.5 mM Ada + 0.5 mM β-CD), 0.25 mM, 0.1 mM, 0.05 mM, and 0.01 mM. Exact formulations including polymer concentration and approximate number of occupied PEG arms are included in Supplementary Table 1. All concentrations demonstrated viscoelastic behavior over the entire frequency range (Figure 7A). We found that the effect of the intra-particle guest-host interactions reached a maximum storage moduli with the combined effect of both the contribution of the guest-host interaction without the loss of polymer concentration and the presence of covalent crosslinks ([0.01 mM] = 910 Pa, [0.05 mM] = 1570 Pa, [0.1 mM] = 1690 Pa, [0.25 mM] = 540 Pa, [0.5 mM] = 530 Pa).

Figure 7: Rheology of mono-MAP-PEG concentrations.

(A) Frequency sweeps from 10 to 0.01 Hz at 1 % amplitude for mono-MAP-PEG hydrogels with [0.5 mM], [0.25 mM], [0.1 mM], [0.05 mM], and [0.01 mM] of Adamantane and β-cyclodextrin. (B) Oscillatory strain sweep from 100 % to 0.001 %. (C) Cyclic strain alternating between low (1 %) strain for 120 s and high (500 %) strain for 60 s over 3 cycles. (D) Unidirectional shear rate sweeps from 100 1/s to 0.01 1/s. (E) Calculated yield stress from the unidirectional shear rate sweep (D). Significance between means of three or more groups determined by one-way ANOVA, p < 0.05 with Tukey’s post-hoc pairwise comparisons. ** = p < 0.01, *** = p < 0.001.

We then conducted an oscillatory strain sweep from 0.001 to 500% at a constant frequency and determined an approximate yield stress for each concentration of guest-host molecules on Intra-MAP-PEG ([0.5 mM] = 120 Pa, [0.25 mM] = 90 Pa, [0.1 mM] = 470 Pa, [0.05 mM] = 500 Pa, [0.01 mM] = 170 Pa). We investigated the shear-thinning and self-healing properties of the different concentrations of Intra-MAP-PEG through a cyclic strain sweep. We alternated between periods of low (1 %) and high (500 %) strain to measure the material response between the two periods. We found that all concentrations were capable of shear thinning and self-healed after a period of high strain (Figure 7C). There were no significant differences between the rates of recovery between the different concentrations. We conducted a unidirectional shear rate sweep from 100 to 0.01 Hz to further assess the differences in yield stress between the different concentrations (Figure 7D). A linear regression was used to determine the yield stress of each flow curve (Figure 7E). We found that the 0.05 mM and 0.1 mM had significantly greater yield stress than the 0.01 mM, 0.25 mM, and 0.5 mM groups ([0.01 mM] = 70 Pa, [0.05 mM] = 300 Pa, [0.1 mM] = 260 Pa, [0.25 mM] = 30 Pa, [0.5 mM] = 30 Pa). We found that there is a Goldilocks zone for the concentration of guest-host interactions that can be included in the MAP-PEG hydrogel structure before the effect of decreasing covalent crosslinks affects the hydrogel mechanical properties.

Discussion

Here we investigated the effect of inter- and intra- particle guest-host interactions on the mechanical properties of PEG-MAL granular hydrogels composed of two or one species of PEG microgel. Previously we have shown the benefits of using granular hydrogels for rapid cell migration and the advantages of using spherical microgels for cell and islet delivery systems (7, 10). In this study we expanded our understanding of guest-host interactions, their balance between reversible and covalent interactions, shear thinning behaviors, and yield stress.

MAP-PEG scaffolds with guest-host interactions were formed either with inter-particle guest-host interactions or intra-particle guest-host interactions. Inter-MAP-PEG was created using two separate species of PEG microgels: Ada-PEG Microgels (guest), β-CD-PEG Microgels (host), whereas Intra-MAP-PEG was a homogeneous microgel system made of Ada-β-CD-PEG Microgels (Figure 1B, C). We found that for Inter-MAP-PEG when we mixed the Ada-PEG microgels and the β-CD-PEG microgels they demonstrated a well mixed scaffold. We controlled for the mixing of two species of gel in the Inter-MAP-PEG by generating two species of Intra-MAP-PEG and MAP-PEG labeled with two different dyes and mixing them together to study the hydrogel structure. We found that there were no significant differences in the mixing and overall structure of the hydrogel scaffolds, indicating that the presence of guest-host interactions does not have a significant effect on the hydrogel structure.

To determine if guest-host interactions were occurring between microgels, we incubated Inter-MAP-PEG, Intra-MAP-PEG, and MAP-PEG with β-CD-Alexa Fluor 647. Inter-MAP-PEG and Intra-MAP-PEG showed significantly more bound β-CD than MAP-PEG, as determined by Pearson’s coefficient and MFI, indicating that guest-host molecules were able to interact on the surfaces of the microgels (Figure 3). We further confirmed the presence of guest-host interactions on the MAP-PEG bulk hydrogels (Figure 4B). We found that both the Intra-MAP-PEG and Inter-MAP-PEG precursors could adhere, whereas the control β-CD-PEG to β-CD-PEG was unable to adhere. This behavior suggests that with greater surface area brought into contact, the guest-host interactions have an additive effect that increases the strength of the interaction.

We investigated the role of guest-host interactions on the material properties of PEG microgels through rheological studies. Inter-MAP-PEG, Intra-MAP-PEG, and unfunctionalized MAP-PEG all exhibited solid-like behaviors at low strains, and frequency-independent elastic moduli (Figure 5A). At 1 Hz, the elastic modulus of Inter-MAP-PEG and unfunctionalized MAP-PEG were significantly greater from the Intra-MAP-PEG, furthermore, the yield stress as measured in a unidirectional flow shear rate sweep were also significantly greater for the Inter-MAP-PEG and MAP-PEG hydrogels (Figure 5E). This was contrary to our initial hypothesis that intra-particle guest-host interactions would maintain the same mechanical properties of dual-MAP-PEG with a more homogeneous structure. We further explored this hypothesis by assessing the polymer concentration of the hydrogel scaffold by altering the weight percentage of the microgels while keeping the guest-host concentration constant (Figure 6). We found that Intra-MAP-PEG had a significantly lower yield stress for all weight percentages than the Inter-MAP-PEG and MAP-PEG scaffolds (Figure 6E). This led us to believe that the concentration of the guest-host interactions was a large contributing factor to the mechanical properties of the hydrogel. We then explored this hypothesis by titrating the concentration of guest-host molecules on the Intra-MAP-PEG and measuring the resulting mechanical properties (Figure 7). We found that there was a maximum concentration of guest-host molecules that could be included in the PEG-MAL structure before the presence of functional groups on the polymer chain reducing covalent crosslinking, had an effect. We found this concentration to be between 0.05 mM and 0.1 mM of guest-host molecules on the polymer backbone (Figure 7E). At these concentrations, we found that the covalent crosslinking and the reversible guest-host crosslinking had a synergistic effect and increased the yield stress and modulus of the hydrogel three times greater than the unfunctionalized MAP-PEG.

Conclusions

In this study, we investigated the effect of inter and intra particle guest-host interactions on the mechanical properties of PEG-MAL granular hydrogels. We demonstrated that Intra-MAP-PEG hydrogels with Intra-GH crosslinking resulted in homogeneous microgel structure with a consistent diameter. Furthermore, we show that guest-host molecules on mono-MAP-PEG are available for guest-host interactions, and that these guest-host interactions have a decreasing effect on the rheological and mechanical properties of the hydrogel, resulting in soft, yielding hydrogels. We altered different hydrogel properties such as wt % and concentration of guest-host interactions and found that Intra-MAP-PEG has increasing mechanical properties with increasing polymer wt %. We found that there was a limit to the concentration of guest-host interactions that could be added to the Intra-MAP-PEG structure before further addition of guest-host molecules started to decrease the mechanical properties.

Methods

Chemicals and Reagents:

4-arm PEG-MAL (PEG-4MAL, 20kDa) was purchased from Laysan Bio. 4-arm PEG-thiol (PEG-4SH, 20kDa) was purchased from Jenkem Technology. 1-adamantane-thiol, Span 80, mineral oil, triton-X 100, 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid (HEPES), dimethyl sulfoxide (DMSO) were obtained from Sigma Aldrich. Mono(6-mercapto-6-deoxy)-β-cyclodextrin was obtained from Zhiyuan Biotechnology.

Microsphere Synthesis:

For mono-MAP-PEG the PEG-4MAL (20kDa) macromer was dissolved in 1X phosphate buffered saline (PBS) with 1% HEPES at pH 5.4 (70.8 mg/mL, 1.7 mM). The reduced pH helps to slow down the crosslinking reaction which at neutral pH is so quick it hinders mixing and handling. (30) At pH 5.4 the gelation of PEG-4MAL with PEG-4SH takes approximately 34 seconds. (7) Adamantane-thiol was dissolved in 1X PBS with 1 % HEPES and 10 % DMSO at pH 5.4 (0.84 mg/mL 0.25 mM) and mono-6-mercapto-β-cyclodextrin was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (5.6 mg/mL, 0.25 mM). Adamantane-thiol and mono-6-mercapto-β-cyclodextrin were added dropwise to the PEG-4MAL macromer at 1 Ada:1 β-CD molar ratio and reacted for 30 minutes at 25°C. PEG-4SH (20 kDa) was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (48.4 mg/mL, 1.2 mM) and reacted with a trace amount of Alexa Fluor-MAL for 5 minutes at 25°C to aid in microgel visualization. PEG-4SH was added to the functionalized PEG-4MAL macromer at a 1:1 volume ratio, quickly pipetted up and down several times to mix thoroughly, then transferred to a 30x volume of mineral oil with 2 % vol/vol Span 80 surfactant in a 50 mL conical tube. The tube is immediately vortexed for 30 s to generate an emulsion, then allowed to finish crosslinking for 30 minutes at 25°C on a rocker plate to generate microgels of final polymer concentration of 6 % wt/vol.

For dual-MAP-PEG the PEG-4MAL (20 kDa) macromer was dissolved in 1X PBS with 1% HEPES at pH 5.4 (70.6 mg/mL, 1.7 mM) and split into two equal volumes. Adamantane-thiol was dissolved in 1X PBS with 1 % HEPES and 10% DMSO at pH 5.4 (1.68 mg/mL, 0.5 mM) and mono-6-mercapto-β-cyclodextrin was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (11.2 mg/mL, 0.5 mM). Adamantane-thiol and β-cyclodextrin were added dropwise to separate PEG-4MAL macromer aliquots and reacted for 30 minutes at 25°C to create to separate species of PEG-4MAL macromers. PEG-4SH was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (48.7 mg/mL, 1.2 mM) and reacted with a trace amount of Alexa Fluor-MAL for 5 minutes. PEG-4SH was added to the separate PEG-4MAL macromers at a 1:1 volume ratio, and microgels were formed by emulsion the same as above.

For MAP-PEG the PEG-4MAL (20 kDa) macromer was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (70.8 mg/mL, 1.4 mM). PEG-4SH (20 kDa) was dissolved in 1X PBS with 1 % HEPES at pH 5.4 (59.14 mg/mL, 1.4 mM) and reacted with a trace amount of Alexa Fluor-MAL for 5 minutes. PEG-4SH was added dropwise to the PEG-4MAL macromer at a 1:1 volume ratio, and microgels were formed by emulsion the same as above.

For all microspheres, after crosslinking the microgels were centrifuged at 3,000 x g for 5 minutes and washed three times with 0.3 % Triton X-100 in 1X PBS at pH 7.2, once with 50 % acetone in DI water, once with DI water, and finally once with 1X PBS at pH 7.2. After washing procedures, the two species of Adamantane-PEG-4MAL and the β-cyclodextrin-PEG-4MAL microgels were added to a single 50 mL tube and vortexed for 30 minutes to form the dual-MAP-PEG.

Guest-Host Interaction Confirmation:

Mono-MAP-PEG, dual-MAP-PEG, and MAP-PEG were formed as described above. Mono-6-mercapto-β-cyclodextrin was dissolved in 1X PBS at pH 7.2 at 3 mM and incubated with an equimolar amount of Alexa Fluor-Maleimide (488, 568, or 647). Microgels were incubated with labeled Alexa-Fluor-β-cyclodextrin for 1 hour. Post-incubation, the microgels were washed 2 times with 1X PBS and imaged for fluorescence intensity of the labeled Alexa Fluor-β-cyclodextrin on Intra-MAP-PEG, Inter-MAP-PEG, and MAP-PEG. For the Inter-MAP-PEG the colocalization of the dye on the Adamantane-PEG was compared to the colocalization of β-Cyclodextrin-PEG colocalization. Colocalization was measured by Pearson’s Coefficient using the JACoP plug-in for ImageJ. (31)

Force of Injection Measurements:

Force of Injection measurements were taken using a force sensor (Tekscan, Flexiforce), Redboard (SparkFun), and syringe pump. The plunger of a luer-lock 1 ml syringe (BD) was removed, using a spatula, 200 μl microgels were loaded into the back of the syringe. The plunger was replaced, and the gels were compressed into the bore of the syringe. A 1” 20 G needle (BD Precision Glide) was connected, and the syringe was then loaded onto a syringe pump with the force sensor placed between the syringe plunger and pump actuator. The force sensor was connected to a Redboard for data acquisition. The microgels were extruded for 30 seconds at a volumetric flow rate of 600 μl minute⁻¹. Voltage output was recorded using Arduino IDE. The extrusion force was calculated by converting voltages to forces using a standard force-voltage calibration curve created using known weights before experimentation. The wall shear stress was determined by using the derived Navier Stokes equations of a turbulent shear-thinning fluid through the needle of length ($L$) and internal diameter ($d$) due to a pressure drop ($\Delta P$) within the needle for microgels and fragmented gels (Equation 2). The pressure drop was defined as the difference between the entrance pressure applied at the plunger and the exit pressure at the needle opening.

$${\tau }_w=\frac{\Delta P\mathrm{*}R}{2L}$$ (1)

Rheological Measurements:

Rheological measurements were performed on an Anton Paar MCR 702 rheometer with a 20 mm sandblasted plate on plate configuration with 0.5 mm gap height at 25°C. To load samples on the rheometer, about 1 mL of the microgels were placed on the bottom plate at room temperature. Oscillatory shear strain amplitude sweeps were performed at 1 Hz between 0.01 % and 500 % strain. The yield stress and strain were determined by taking the second derivative of the fitted curve and finding the point of inflection. The storage (G’) and loss modulus (G”) were determined by taking 2-minute time sweeps at 1 % strain and 1 Hz frequency. Unidirectional shear rate sweeps were performed by shearing the samples at shear rates (γ) ranging from 100 1/s to 0.01 1/s and while measuring shear stress (σ). A linear regression was performed to determine the zero-frequency limit and resultant yield stress of the microgels on GraphPad. Strain cycle and recovery experiments were conducted by alternating 1 % strain for 120 seconds and 500 % strain for 60 seconds over three periods at 1 Hz.

Microscopy:

Microgels were imaged on a Leica SP8 confocal laser-scanning microscope using 10x/0.3 and 20x/0.8 numerical aperture Plan-Apochromat air objectives at 1,024 × 1,024-pixel resolution. Images were processed and quantified using the FIJI distribution of ImageJ.(32)

Statistical Analysis:

Means among three or more groups were compared by a one-way analysis of variance (ANOVA) in GraphPad Prism 8 software. If deemed significant, Tukey’s posthoc pairwise comparisons were performed. Means between two groups were compared by two-tailed Student’s t-test. A confidence level of 95% was considered significant. The statistical test used, exact P values, and definition of n are all indicated in the individual figure legends. All error bars in the figures display the mean ± s.e.m.