Bioinspired Oxidation-Resistant Catechol-like Sliding Ring Polyrotaxane Hydrogels

Adaptable hydrogels have been used in the biomedical field to address several pathologies, especially those regarding tissue defects. Here, we describe unprecedented catechol-like functionalized polyrotaxane (PR) polymers able to form hydrogels. PR were functionalized with the incorporation of hydroxypyridinone (HOPO) moieties into the polymer backbone, with a degree of substitution from 4 to 22%, depending on the PR type. The hydrogels form through the functionalized supramolecular systems when in contact with a Fe(III) solution. Despite the hydrogel formation being at physiological pH (7.4), the HOPO derivatives are extremely resistant to oxidation, unlike common catechols; consequently, they prevent the formation of quinones, which can lead to irreversible bounds within the matrix. The resulting hydrogels demonstrated properties lead to unique hydrogels with improved mechanical behavior obtained by metallic coordination crosslinking, due to the synergies of the sliding-ring PR and the non-covalent (reversible) catechol analogues. Following this strategy, we successfully developed innovative, cytocompatible, oxidative-resistant, and reversible crosslinked hydrogels, with the potential of being used as structural self-materials for a variety of applications, including in the biomedical field.


Introduction
Several works have explored the possibility of mimicking the byssus threads of mussels, such as Mytilus edulis. Studies have found that this type of mollusk shows strong adhesion ability to rocks in sea conditions, that is, in wet environments and with external force against the attachment. This phenomenon is possibly due to the adhesive properties present in the mussel's secreted protein that contains the 3,4-dihydroxy-L-phenylalanine (DOPA) amino acid precursor in its structure, which is the key factor in the adhesion capacity of mussels [1]. Due to the high versatility, good biocompatibility, and excellent adhesion performance, mussel-inspired catechol strategies have been widely explored through chemical modification of several molecules, such as synthetic and natural polymers, to improve wet adhesion, adherence to surfaces, and self-healing capacities [2][3][4][5][6][7]. However, the easy and fast oxidation of these entities requires the use of acidic conditions for the manipulation of these compounds, which are not optimal when biomedical applications are intended [8].
Regarding this field, a plethora of polymeric-based systems have been developed due to their similarity to the extracellular matrix, such as nanofibers [9] or several class of composite hydrogels, combining the features of two or more polymeric systems to address the typical issues experienced with polysaccharides [10]. In particular, polymeric hydrogels have been widely proposed as artificial matrices to support new tissue development, for

Preparation and Characterization of PR-HOPO Hydrogels
To evaluate the hydrogel formation ( Figure 3), polymers 7a-d were dispersed in MES buffer at pH 5.6 and 10% (w/v), and were prepared following the concentrations proposed by Holten-Andersen et al. [15], with slight modifications. Final hydrogels were composed by 1/2 of the volume of the dispersion of 7a-d, 1/3 of the volume of the metal solution (20 mM FeCl 3 ), and 1/6 of the volume of NaOH (0.1 M), in order to increase the pH of the medium to ca. 7.4 to promote the tris-catecholate complex between the metal and HOPO moieties [33]. The hydrogel formation was observed almost immediately after pH changing (Video S1). A semisolids pH meter was used to monitor the pH throughout the whole process, avoiding a pH greater than 7.4. All developed systems were demonstrated to be robust and stable in PBS solution at pH 7.4 for 24 h. The hydrogels showed soft adhesion properties and an interesting capacity to self-heal after a scratch (Video S2). Gels 2023, 9, x FOR PEER REVIEW 4 of 11 groups (6d, Figure 2B), were functionalized with the prepared HOPOs (4a or 4b) to avoid quick oxidation, an issue widely reported for traditional catechol moieties [14,30]. To do so, PR 6a-c and 6d were coupled with HOPOs 4a and 4b, respectively, by EDC/NHS coupling reaction in MES buffer at pH 5.6. After a 12 h reaction period at room temperature (RT), the products were dialyzed and freeze-dried. All PR-HOPO (7a-d) polymers were obtained as white/slightly yellow solids, with a substitution degree (DS) of 4%, 9%, 22%, and 15% (w/w), respectively, confirmed by 1 H NMR and quantified by UV-Vis spectroscopy (see 1.2, ESI) [29,30]. Comparing the 1 H NMR spectra of the respective HOPO precursor (4a or 4b), PRs 6a-d, and functionalized PRs with HOPO 7a-d, it was possible to identify two new doublets around 6.52 ppm (J = 7.1 Hz, i.e., for 7c) and 7.59 ppm (J = 7.1 Hz, i.e., for 7c), the characteristic signals corresponding to protons H-5 and H-6 from the HOPO precursor, respectively, which confirmed the HOPO moieties' insertion into the PRs chain ( Figures S5-S8).

Preparation and Characterization of PR-HOPO Hydrogels
To evaluate the hydrogel formation ( Figure 3), polymers 7a-d were dispersed in MES buffer at pH 5.6 and 10% (w/v), and were prepared following the concentrations proposed by Holten-Andersen et al. [15], with slight modifications. Final hydrogels were composed by 1/2 of the volume of the dispersion of 7a-d, 1/3 of the volume of the metal solution (20 mM FeCl3), and 1/6 of the volume of NaOH (0.1 M), in order to increase the pH of the medium to ca. 7.4 to promote the tris-catecholate complex between the metal and HOPO moieties [33]. The hydrogel formation was observed almost immediately after pH changing (Video1_SI2). A semisolids pH meter was used to monitor the pH throughout the whole process, avoiding a pH greater than 7.4. All developed systems were demonstrated to be robust and stable in PBS solution at pH 7.4 for 24 h. The hydrogels showed soft adhesion properties and an interesting capacity to self-heal after a scratch (Video2_SI3). The PR-HOPO derivatives 7a-c led to hydrogels with elastic modulus (1 kPa) greater than the already reported ones modified with traditional catechol moieties [17]. In all The PR-HOPO derivatives 7a-c led to hydrogels with elastic modulus (1 kPa) greater than the already reported ones modified with traditional catechol moieties [17]. In all cases, the storage modulus was greater than the loss modulus, which is a classic behavior of gellike materials [34], and no changes in the moduli appeared after the cut (Figures 4 and S9). This behavior confirmed the hypothesis of reversible hydrogels formation, just by metallic coordination.
functionalized with amino (6d) and carboxylic groups in several degrees (6a was less substituted and 6c was more substituted; Figure 2B). Then, the obtained insertion of HOPO groups was performed according to the initial available amount of amino or carboxylic groups of 6a-d, i.e., 7a had the smallest DS and 7c the higher one. Consequently, PR-HOPO 7b and 7c demonstrated a greater elastic modulus than PR-HOPO 7a ( Figure S10). Additionally, by the DS of HOPO groups in the PRs chain, it was possible to determine the number of possible metallic coordinations that these supramolecular systems can form. Thus, we could produce mechanically versatile hydrogels that are only dependent on the availability of groups to take part in the metallic coordination. This is especially important and opens a way to new approaches for the development of structural hydrogels, for example, for engineering substitute tissues and potential tissue substitutes, retaining the properties of the native ones.

Assessment of Cell Viability upon Contact with PR-HOPO Formulations
The cytocompatibility of the hydrogels was assessed with SAOS-2 cells ( Figure 5). The extracts from the samples were performed under the guidelines of ISO standard 10993:5 (cytotoxicity of medical devices). After 24 and 48 h, a cell proliferation test (MTS) was carried out following the instructions of the manufacturer. Hydrogels based on 7a, 7b, and 7c showed a 70% proliferation when compared with controls in all formulations, classified as non-cytotoxic according to ISO 10993:5. Statistical analysis demonstrated that formulation 7d differed from the other samples, with lower values of cell proliferation (confidence level of 95%, significance level (α) of 0.05). On the other hand, there were no statistical differences in the viability of the cells for the same sample at 24 and 48 h. Commonly, for the crosslinking of these types of network, the pH must be increased to around 8-9 [15]. Due to the needs of this pH, we could hypothesize the non-compatibility of the systems. In our case, we kept complete control by checking the pH from the synthesis of the PR-HOPO until the metallic coordination, maintaining it at 5.6 during the reaction process to 7.4 to promote the metallic coordination. Moreover, a clear difference between the four formulations was observed, according to the substitution degree of each PR. Another important parameter that should be taken into account is the number of CDs along the PEG chain [35], because this is directly related to the flexibility and elasticity of the system. In this study, as a proof of concept, all PRs contained 100 CDs for each PEG chain (Mw 35,000 Da). As mentioned above, the PRs were functionalized with amino (6d) and carboxylic groups in several degrees (6a was less substituted and 6c was more substituted; Figure 2B). Then, the obtained insertion of HOPO groups was performed according to the initial available amount of amino or carboxylic groups of 6a-d, i.e., 7a had the smallest DS and 7c the higher one. Consequently, PR-HOPO 7b and 7c demonstrated a greater elastic modulus than PR-HOPO 7a ( Figure S10). Additionally, by the DS of HOPO groups in the PRs chain, it was possible to determine the number of possible metallic coordinations that these supramolecular systems can form. Thus, we could produce mechanically versatile hydrogels that are only dependent on the availability of groups to take part in the metallic coordination. This is especially important and opens a way to new approaches for the development of structural hydrogels, for example, for engineering substitute tissues and potential tissue substitutes, retaining the properties of the native ones.

Assessment of Cell Viability upon Contact with PR-HOPO Formulations
The cytocompatibility of the hydrogels was assessed with SAOS-2 cells ( Figure 5). The extracts from the samples were performed under the guidelines of ISO standard 10993:5 (cytotoxicity of medical devices). After 24 and 48 h, a cell proliferation test (MTS) was carried out following the instructions of the manufacturer. Hydrogels based on 7a, 7b, and 7c showed a 70% proliferation when compared with controls in all formulations, classified as non-cytotoxic according to ISO 10993:5. Statistical analysis demonstrated that formulation 7d differed from the other samples, with lower values of cell proliferation (confidence level of 95%, significance level (α) of 0.05). On the other hand, there were no statistical differences in the viability of the cells for the same sample at 24 and 48 h. Commonly, for the crosslinking of these types of network, the pH must be increased to around 8-9 [15]. Due to the needs of this pH, we could hypothesize the non-compatibility of the systems. In our case, we kept complete control by checking the pH from the synthesis of the PR-HOPO until the metallic coordination, maintaining it at 5.6 during the reaction process to 7.4 to promote the metallic coordination. Tests were carried out in triplicate. * Statistical differences were found in formulation 7d when compared with the rest of the formulations.

Conclusions
Overall, biocompatible, cutting-edge, oxidation-resistant, and self-healing hydrogels were developed, taking advantage of functionalized sliding-ring PRs with HOPO moieties and avoiding the common oxidation issues. This completely novel macromolecular design results in hydrogels with the capacity to distribute the typical stresses among the polymer chains subject to compression stress, along the entire volume of the hydrogel through reversible free movements of CDs. It is important to note that the developed hydrogels were obtained with just one network; yet, they showed interesting mechanical properties. The possibility to modulate the crosslinking degree and tune the properties of the hydrogels makes this approach a versatile system. Here, we proved that PR-HOPO formulations are able to create reversible systems just by coordination crosslinking. The capacity to maintain the polymeric network integrity when the coordination of the HOPO is achieved at physiological pH makes them compatible with osteoblast cells. These features give an added value to these systems, markedly differentiated from the existing systems based on the same chemistry.

Conclusions
Overall, biocompatible, cutting-edge, oxidation-resistant, and self-healing hydrogels were developed, taking advantage of functionalized sliding-ring PRs with HOPO moieties and avoiding the common oxidation issues. This completely novel macromolecular design results in hydrogels with the capacity to distribute the typical stresses among the polymer chains subject to compression stress, along the entire volume of the hydrogel through reversible free movements of CDs. It is important to note that the developed hydrogels were obtained with just one network; yet, they showed interesting mechanical properties. The possibility to modulate the crosslinking degree and tune the properties of the hydrogels makes this approach a versatile system. Here, we proved that PR-HOPO formulations are able to create reversible systems just by coordination crosslinking. The capacity to maintain the polymeric network integrity when the coordination of the HOPO is achieved at physiological pH makes them compatible with osteoblast cells. These features give an added value to these systems, markedly differentiated from the existing systems based on the same chemistry. meter with a penetration probe for semisolid systems (Dominique Dutscher, Bernolsheim, France). The UV-Vis spectra and MTS assays were recorded on a Synergy HTX Microplate Reader (Agilent Technologies, Santa Clara, CA, USA), using a 96-well quartz microplate. Rheological tests were performed with a Kinexus Lab+ rheometer (Malvern Panalytical, Malvern, UK), by using a 20 mm diameter parallel plate geometry and a 0.5 mm gap at 25 • C. To determine the linear viscoelastic region (LVR), strain amplitude sweep measurements (0.1 to 1000%) were performed at a frequency of 1.0 Hz. Oscillatory frequency sweep measurements (0.01 to 100 Hz) were then conducted at constant 1% strain amplitude to determine the storage (G ) and loss (G") moduli.

1-(2-aminoethyl)-3-hydroxy-2-methylpyridin-4(1H)-one (4a).
Compound 4a was synthesized as follows, based on a previously reported method [29,30]. Maltol (1) (5.0 g, 39.6 mmol, 1.0 eqv) was added to hot water (100 mL) in the presence of EDA (2) (4.0 mL, 59.4 mmol, 1.5 eqv). To promote total dissolution, the mixture was stirred and heated to 80 • C. The pH solution was adjusted to 10 by dropwise addition of 6 M HCl solution and monitored by pH paper. The mixture was brought to reflux and stirred for 24 h and then cooled to RT. All water was removed under reduced pressure. The crude product was dissolved in 20 mL of distilled water, and its pH was adjusted to~7 with the addition of 6 M HCl solution. All water was again evaporated. The product was precipitated by the addition of methanol/diethyl ether. A light brown precipitate was isolated by filtration, washed with cold methanol, and dried in an oven for 12 h, at 60 • C, to yield compound 4a

2-(3-hydroxy-2-methyl-4-oxopyridin-1-(4H)-yl)acetic acid (4b).
The procedure for the synthesis of compound 4b was adapted from a procedure reported in the literature [29,30]: Maltol (1) (7.5 g, 59.5 mmol, 1.0 eqv) was added to hot water (150 mL) in the presence of glycine (3) (8.9 g, 119.0 mmol, 2.0 eqv). To promote total dissolution, the mixture was stirred and heated to 80 • C. The pH of the solution was increased to 9 by the dropwise addition of 6 M NaOH solution and monitored by pH paper. The mixture was brought to reflux and maintained for 24 h. The reaction mixture was cooled to room temperature (RT), and almost all water was removed under reduced pressure using a rotary evaporator. The pH of the crude product was brought to~3 with the addition of a 6 M HCl solution, at which time a beige solid precipitated was formed. The precipitate was isolated by filtration and was subsequently recrystallized from hot water and stored at 4 • C overnight. The resulting solid was filtered off, washed with distilled water, and dried in an oven at 60 • C for 12 h, to yield a beige solid, which was identified as compound 4b

General Procedure for Polyrotaxane-Hydroxypyridinones (PR-HOPO) Functionalization
HOPO-NH 2 (4a) and HOPO-CO 2 H (4b) were conjugated onto the PR-CO 2 H (6a-c) and PR-NH 2 (6d) chains, respectively, by EDC/NHS coupling reactions. Thus, PR-CO 2 H or PR-NH 2 (100 mg, 1 eqv) was dissolved in 50 mL of MES buffer at pH 5.6 into a reaction vial. Then, EDC (287.51 mg, 6.0 eqv) was dissolved in 10 mL of MES buffer at pH 5.6 and added to the previous reaction vial for 10 min. Likewise, NHS (172.63 mg, 6.0 eqv) was dissolved in 10 mL of MES buffer pH 5.6 and added to the reaction mixture. After 15 min, HOPO-NH 2 (4a) or HOPO-CO2H (4b) (50.74 mg or 55.85 mg respectively, 6.0 eqv), Gels 2023, 9, 85 8 of 11 previously dissolved in 10 mL of MES buffer pH 5.6, was added to the previous reaction mixture and left to react at RT for 12 h. Then, the resulting reaction solution was transferred to dialysis bags and was dialyzed against MES buffer pH 5.6 for 5 days under stirring at RT. The MES buffer was changed twice per day. PR-HOPO derivatives (7a-d) were obtained by freeze-drying as a pale white/slightly yellow solid and stored at −20 • C.

Determination of the Degree of Substitution (DS)
The different degrees of substitution of HOPOs in the PRs were determined by UV-Vis spectroscopy at 276 nm (standard curve from 1 to 20 µg·mL −1 of HOPO-NH 2 (4a) in ultrapure water: y = 0.05343x − 0.0062, R 2 = 0.9991; standard curve of HOPO-CO 2 H (4b) in ultrapure water: y = 0.0551x − 0.0061, R 2 = 0.9930). Solutions of 20 µg·mL −1 in ultrapure water of 7a-d were prepared, and the absorbance at 276 nm wavelength was compared with the standard curves [29,34]. Solutions of each PR at the same concentration (20 µg·mL −1 ) were used as blanks. All measurements were conducted in triplicate.

Preparation and Characterization of HOPO-PR Hydrogels
Polymers 7a-d were dispersed in MES buffer at pH 5.6 and 10% (w/v) to prepare the hydrogels, following the concentrations proposed by Holten-Andersen et al. [16]. The final hydrogels were composed by 1/2 of the volume of the dispersion of 7a-d, 1/3 of the volume of the metal solution (20 mM FeCl 3 ), and 1/6 of the volume of NaOH (0.1 M), in order to increase the pH medium at 7.4 ca. to promote the tris-catecholate complex between the metal and HOPO moieties [36]. The pH of the hydrogels was monitored during the formation with a pH meter with a penetration probe for semisolid systems (Dutscher).
The elastic and the viscous moduli and the self-healing ability of prepared hydrogels were evaluated. The mechanical properties of the PR-HOPO-based hydrogels before and after a scratch were established by performing frequency and strain sweep tests. Rheological tests were performed with a Kinexus Lab+ rheometer (Malvern Panalytical), by using a 20 mm diameter parallel plate geometry and 0.5 mm gap at 25 • C. To determine the linear viscoelastic region (LVR), strain amplitude sweep measurements (0.1 to 100%) were performed at a frequency of 1.0 Hz. Oscillatory frequency sweep measurements (0.01 to 100 Hz) were then conducted at constant 1% strain amplitude to determine the storage (G ) and loss (G") moduli.

Assessment of Cell Viability upon Contact with PR-HOPO Formulations
In order to establish the cytocompatibility of the PR-HOPO formulations, cell proliferation MTS tests were carried out after contact with the extracts of PR-HOPO with osteoblast SAOS-2 cells. Firstly, extracts from the samples were obtained as described in ISO standard 10993:5 and 10993:12 guidelines. A total of 500 mg of each sample was extracted in triplicate at 37 • C during 24 h in DMEM culture (1 mL). Then, 100 µL of SAOS-2 osteoblast cells suspension (concentration 200,000 cells mL −1 ) was seeded in 96-well plates. After 24 h of incubation at 37 • C, 95% relative humidity (RH), and 5% CO 2 , several concentrations (50%, 25%, 12.5%, and 6.25% v/v) extracted from the samples were added and placed in Gels 2023, 9, 85 9 of 11 contact with the cells for 24 and 48 h. Then, a cell proliferation test (MTS Proliferation Assay Kit ab197010, Abcam) was carried out following the instructions of the manufacturer. As controls, cells seeded with complete culture medium only were used. Absorbance was measured at 490 nm on a Synergy HTX Microplate Reader using a 96-well quartz microplate. All experiments were carried out in triplicate.

Statistical Analysis
Results are shown as mean values and standard deviations of at least three replicates. Regression analysis for the determination of the degree of substitution was carried out using Microsoft Excel Office 365 software (Microsoft Corporation, Redmond, WA, USA). Differences in the types of PR-HOPO formulations regarding the cell proliferation were determined by applying a two-way ANOVA test (confidence level of 95%, significance level (α) of 0.05) and Tukey's multiple comparison test with GraphPad Prism 8.3.0 (GraphPad Software Inc, New York, NY, USA).