Self-Healing Dynamic Hydrogel as Injectable Shock-Absorbing

Jul 5, 2017 - Herein, we evaluate an injectable, self-healing dynamic hydrogel (DH) based on gold(I)-thiolate/disulfide (Au–S/SS) exchange as NP rep...
0 downloads 12 Views 2MB Size
Subscriber access provided by UNIVERSITY OF CONNECTICUT

Article

Self-Healing Dynamic Hydrogel as Injectable Shock-Absorbing Artificial Nucleus Pulposus Adrián Pérez-San Vicente, Marianna Peroglio, Manuela Ernst, Pablo Casuso, Iraida Loinaz, Hans-Jurgen Grande, Mauro Alini, David Eglin, and Damien Dupin Biomacromolecules, Just Accepted Manuscript • DOI: 10.1021/acs.biomac.7b00566 • Publication Date (Web): 05 Jul 2017 Downloaded from http://pubs.acs.org on July 6, 2017

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a free service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are accessible to all readers and citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

Biomacromolecules is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Self-Healing Dynamic Hydrogel as Injectable Shock-Absorbing Artificial Nucleus Pulposus Adrián Pérez-San Vicente,a,‡ Marianna Peroglio,b,‡ Manuela Ernst,b Pablo Casuso,a Iraida Loinaz,a* Hans-Jürgen Grande,a Mauro Alini,b David Eglin,b Damien Dupina* a

Materials Division, IK4-CIDETEC Research Centre, Paseo Miramón 196, DonostiaSan Sebastián 20014, Spain. b

AO Research Institute Davos, Clavadelerstrasse 8, 7270 Davos, Switzerland.

KEYWORDS: Dynamic hydrogels, Self-Healing, Shock Absorbing, Intervertebral Disc, Nucleus Pulposus.

ABSTRACT. The intervertebral discs (IVDs) provide unique flexibility to the spine and exceptional shock absorbing properties under impact. The inner core of the IVD, the nucleus pulposus (NP) is responsible for this adaptive behavior. Herein we evaluate an injectable, self-healing dynamic hydrogel (DH) based on gold(I)-thiolate/disulfide (AuS/SS) exchange as NP replacement in a spine motion segment model. For the first time we report the application of dynamic covalent hydrogels inside biological tissues. The dynamic exchange between Au-S species and disulfide bonds (SS) resulted in selfhealing ability and frequency-dependent stiffness of the hydrogel, which was also confirmed in spine motion segments. Injection of preformed DH into nucleotomized IVDs restored the full biomechanical properties of intact IVDs, including the stiffening 1 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

effect observed at increasing frequencies which cannot be achieved with conventional covalent hydrogel. DH has the potential to counteract IVD degeneration associated with high frequency vibrations. Self-healing properties, confirmed by rheology studies and macroscopic observation after injection, were required to inject preformed DH which recovered its mechanical integrity and microstructure to act as an artificial NP. On the other hand, covalent hydrogel did not show any restoration of NP properties as this conventional material suffered irreversible damages after injection, which demonstrates that the dynamic properties are crucial for this application. The persistence of DH in the IVD space following cyclic high-frequency loading, confirmed by tomography after mechanical testing, suggests that this material would have long life span as injectable NP replacement material.

INTRODUCTION Lower back pain (LBP) affects four out of five people in their lifetime and has an important socio-economic impact.1-2 Most LBP cases have been linked with the degeneration of the intervertebral disc (IVD).3-4 IVD, which is located between the vertebras, is composed of two main regions, an inner, soft and highly hydrated structure, called nucleus pulposus (NP), and an outer, firm, collagenous annulus fibrosus (AF) consisting of concentric lamellae encircling NP. NP and AF are naturally designed to maintain the normal biomechanical function of the disc with complementary roles in sharing loads occurring. The central NP confers shock-absorbing properties to the IVD and allows energy dissipation and resistance to low loads, while the surrounding AF plays a more important role at high loads.5-6 The main cause of IVD degeneration is an increased stress on the fibers of AF due to the dehydration of NP.7-8 This loss of water

2 ACS Paragon Plus Environment

Page 2 of 35

Page 3 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

affects the mechanical properties and subsequently the load sharing synergy with AF, which ultimately contributes to further disc degeneration.5,9-11. Injectable hydrogels have received increasing attention because of their similarity to NP tissue and potential to re-hydrate the IVD of patients suffering IVD degeneration using minimally invasive procedures.12 Injectable hydrogels were shown to restore the biomechanics of the intact IVD to some extent. For instance, Milani et al. have proposed injectable doubly cross-linked synthetic microgels to treat degenerated IVD.13 This material was partially cured before injection and could improve the strain, modulus, toughness and resilience of degenerated IVDs. However, the ability to replicate NP stiffening at increasing frequencies (observed in intact IVDs, but not in nucleotomized discs) has yet to be achieved. Dynamic

covalent

chemistry

(DCC),

a

synthetic

strategy

that

yield

thermodynamically stable products overcoming the entropic cost of self-assembly, has been used to form 3D networks with self-healing properties based on reversible breaking/reformation of chemical bonds under equilibrium conditions to reprocess an reuse polymer blends.14-18 More recently, DCC has been employed to design hydrogels by using cross-linkers containing functional groups allowing those reversible reactions. 19

DCC conferred to the resulting hydrogel self-healing properties and frequency-

dependent stiffness, similar to shock absorbing properties, due to the permanent mobility of the polymeric network via the exchange reaction. Examples involving phenylboronate-salicylhydroxamate reaction,20-22 dual acylhydrazone/disulfide bonds,23 radically reshuffling trithiocarbonate,24 and radical-mediated disulfide fragmentation,25 have been reported but the exchange reaction was only activated by the application of external stimuli, such as pH or UV light. Dynamic hydrogels (DH) exhibiting their peculiar properties at physiological pH, i.e. pH 7.4 and 37°C, were mainly based on 3 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Schiff base reaction involving imine groups,26-30 transesterifcation of boronic acid with diols,31 hydrolytic cleavage of oxime bonds,32 and thiol/disulfide exchange.33-34 It is also worth mentioning that supramolecular hydrogel based on host-guest interaction have shown similar mechanical properties; but such materials are not the main focus of this study.35,36 To the best of our knowledge, the only example of dynamic hydrogel based on DCC used for tissue regeneration has been reported by Tseng et al.27 In this work, the dynamic properties of the hydrogel were based on the reversible Schiff base reaction at physiological pH between the primary amine of glycol chitosan and poly(ethylene glycol) functionalized with benzaldehyde at both ends. This injectable and self-healing hydrogel was successfully used as vehicle for repairing the central nervous system deficit. Unfortunately, all DH reported until now are mainly focused on the self-healing properties conferred by the reversible reaction, but are not taking much advantage of the frequency-dependent stiffness of the resulting material. More importantly, all DCCs occurring at physiological conditions involve time consuming chemistry and advanced technical skills to prepare functional polymers with the desired groups, which will result in a material with great properties but unable to reach market and economic viability. In that sense, the large amount of commercially available polymers containing thiols and/or disulfide groups allows DCC based on thiol-disulfide exchange to be considered as the main solution to achieve cheap and easy dynamic hydrogels at atmospheric conditions. Unfortunately, this exchange reaction suffers the inherent aerial oxidation of thiols into disulfide, preventing its use in time. In our group, we have first demonstrated that thiolate could be efficiently protected by the addition of Gold(I) ions to form gold(I)-thiolate (Au-S) species, without losing the nucleophilic character of thiolates required for the exchange reaction with disulfide.37 The resulting injectable and cytocompatible hydrogels exhibited self-healing abilities with stiffening effect for at 4 ACS Paragon Plus Environment

Page 4 of 35

Page 5 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

least 2 years, thanks to the exchange reaction between Au-S species and disulfide (SS) at biologically relevant pH (Figure 1A).37,38,39 For example, the stiffness of DH based on commercially available 4-arms thiol terminated PEG polymer [(PEGSH)4] crosslinked via Au-S species and SS could be easily tuned by increasing the concentration of polymer and/or decreasing the amount of Au-S species.37,39 It is important to mention that decreasing the amount of Au-S species affected the shock absorber behavior of the resulting hydrogel due to slower Au-S/SS exchange. In the same study, the lack of selfhealing and frequency-dependent stiffness of the control hydrogel based on disulfide bonds only proved that the dynamic properties were clearly conferred by the addition of gold ions and the formation of Au-S species. In addition, faster self-healing behavior and relaxation time of the hydrogel at increasing pH indicated faster exchange reaction which confirmed the efficient protection of the nucleophile character of thiolate by gold(I) ions. Thus, hydrogels exhibiting dynamic properties at physiological conditions could be readily prepared by using commercially available compounds.

5 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

6

ACS Paragon Plus Environment

Page 6 of 35

Page 7 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 1. (A) Schematic representation of the 4 arms thiols terminated poly(ethylene glycol) [(PEGSH)4] dynamic hydrogel containing 20 mol.% Au-S species based on the total amount of sulphur while the remaining one are forming disulfide bonds. (B) Digital photos of the selfhealing ability of the dynamic hydrogel at 30 wt.% after breaking manually the piece of hydrogel in two parts and put in contact for a few seconds, which results in the recovery of the native hydrogel. Frequency sweep data depicting shear elastic modulus, G’, (filled symbols) and shear loss modulus, G’’, (open symbols) for (C) the dynamic hydrogel [DH] and (D) the corresponding covalent hydrogel [CH] at 37°C, prepared in the presence of 20 wt% BaSO4 (red) or without (blue), as contrast agent for µCT imaging. Note that the addition of 20 wt% BaSO4 did not affect significantly the rheological behavior of both hydrogels. Normal force response determined by rheology of (E) DH and (F) CH containing 20 wt% BaSO4 at 37°C and different compression speeds, i.e. 10, 100 and 1000 µm.s-1, showing the strain rate dependent stiffness of DH with the increase of normal force response at faster axial compression speeds. The CIDETEC logo appearing in (B) is used with permission from CIDETEC. Herein, we report for the first time the injection and functional behavior of a preformed dynamic hydrogel (DH) as NP replacement in a whole IVD organ. Here, preformed hydrogels were selected as they may represent a safer option compared to in-situ forming hydrogels, as the latter may experience diffusion of precursor solutions in the surrounding tissue, resulting in reproducibility issues. In this study, DH based on 30 wt% (PEGSH)4 containing 20 mol.% Au-S species (based on the theoretical amount of sulphur) with shock absorbing was injected into nucleotomized IVDs. Cyclic tension/compression tests using bovine tail IVD were carried out at different frequencies to determine the mechanical parameters of IVDs, namely range of motion (ROM), neutral zone length (NZ length), neutral zone stiffness (NZ 7 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

stiffness), and creep displacement. Four different groups of 7 specimens each were considered: i) intact IVD as reference and ii) NP-nucleotomized IVD as negative control, iii) NP-nucleotomized IVD restored with dynamic hydrogel [DH] and iv) NP-nucleotomized IVD restored with the corresponding covalently cross-linked hydrogel [CH] to investigate the role of Au-S/SS to act as NP replacement. The persistence and mechanical integrity of both DH and CH in the IVD space were investigated by tomography after mechanical testing and rheological studies after injection, respectively. It has to be mentioned that in situ forming CH would surely be more suitable for this study compared to pre-formed hydrogel as it will obviously lack self-healing properties. However, the conditions to form CH in situ at 30 wt% appeared to be very challenging and preliminary attempts have shown that CH formed in situ was not sufficiently reproducible to be considered in this study. Therefore, preformed CH was used as control to show that the DH is self-healing, although it might not be the best representation of the performance of an injectable covalent hydrogel system. EXPERIMENTAL PART

Hydrogels preparation For dynamic hydrogels (DH), a 33 wt% thiol terminated 4-arm poly(ethylene glycol) PEG [(PEGSH)4, Mn=10,000 g·mol-1, Sigma-Aldrich) solution in minimal essential medium (MEM) in the presence of 20 wt% of barium sulfate (BaSO4, Sigma-Aldrich) as contrast agent was mixed in 9:1 volume ratio with HAuCl4 (0.25 M in deionized water). The final concentration of DH was 30 wt%. For covalent hydrogels (CH), a 30.12 wt% of (PEGSH)4 solution in minimal essential medium (α-MEM) was reacted in a 220:1 volume ratio with poly(ethylene glycol) diacrylate (PEGDA, Mn = 250 g·mol-1, Sigma-Aldrich) via thioMichael addition in the presence of 20 wt% of barium sulfate as contrast agent during 24 hrs.

8 ACS Paragon Plus Environment

Page 8 of 35

Page 9 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

The final concentration of CH was 30wt%. The amount of PEGDA was adjusted to react with all the free thiols previously determined with the Ellman test. 38

Rheological characterization Rheological measurements were carried out using an AR2000ex (TA Instruments) rheometer using a parallel plate geometry (20 mm diameter acrylic plate). The experiments were conducted at constant temperature, i.e. 37°C. Shear storage and shear loss moduli (G' and G'', respectively) were obtained at constant deformation (1 % strain) with increasing frequency (from 0.01 to 50 Hz). To demonstrate the effect of injection on the hydrogels, the disc firstly characterized as described above was placed in a syringe and injected in a cylindrical mold. Slight pressure was applied to the hydrogel in the mold and the hydrogel was left resting overnight. Then, rheological were again carried out on the sample. The compression test protocol consisted of 5 sinusoidal cycles of 10% of the maximum gap distance at different speeds: 10, 100 and 1000 µm·sec-1 compression speed on hydrogels prepared in a cylindrical mold with dimensions of 0.5 x 2.0 cm (height x diameter). The maximum value of the normal force reached at each speed was represented for 3 measurements (scale bars were within the upper line).

Hydrogel supplementation to nucleotomized spinal motion segments Bovine tails (7-10 months old) obtained from the local abattoir were stored frozen. After thawing, motion segments (n=28) were isolated with a sagittal saw and a partial nucleotomy (defect size = 5 mm diameter) was made through the center of the upper vertebra with a drill and an arthroscopic shaver (Smith & Nephew). This approach was chosen in order to keep the AF intact and focus solely on NP repair; in addition, this approach was selected as clinically relevant due to recent clinical advances in the field with efforts to inject the implants via the 9 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

trans-endplate to avoid further damage of AF during the surgery.40 Vertebral bodies were reinforced with Kirchner wires (1 mm diameter, DePuy Synthes). Approximately 150-200 µl of a preformed hydrogel (dynamic or covalent) was injected with a syringe (gauge inner diameter = 2.3 mm) into the defect, which was sealed with a silicone stopper and a polymethylmethacrylate (PMMA) cement (Beracryl, Troller). Nucleotomized intervertebral discs (IVDs) without hydrogel and intact IVDs were used as negative and positive controls, respectively. The vertebral bodies were potted in PMMA and IVD hydration was maintained by covering specimens with saline-soaked gauze. Samples were divided into 4 groups (n=7/group): (i) Intact IVD (ii) NP-nucleotomized IVD (iii) DH injected into NPnucleotomized IVD and (iv) CH injected into NP-nucleotomized IVD.

Figure 2. Schematic representation of the hydrogel filled motion segment preparation. The preformed hydrogels, containing 20 wt% BaSO4 as contrast agent, were injected with a 10 ACS Paragon Plus Environment

Page 10 of 35

Page 11 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

syringe in the NP cavity previously prepared with a drill and an arthroscopic shaver. The nucleotomized space was sealed with a silicon stopper and poly(methyl methacrylate) (PMMA) to avoid any extrusion of the hydrogels during mechanical testing. The vertebral bodies were reinforced with Kirchner wires and embedded in PMMA (Figure S1 in Supporting Information). The specimens were imaged by µCT and tested mechanically in axial compression/tension load at -60N/45N, respectively, for 5 cycles at 0.1 Hz, 1 Hz and 5 Hz, as shown in the representation at the bottom. Finally, µCT was carried out after mechanical testing to assess any damage or leakage of the implanted hydrogels prior to transversal cut for visual inspection.

Biomechanical testing of spinal motion segments Based on the existing literature a protocol to test the mechanical properties of the IVD was developed.5,41,42The testing protocol consisted of a preconditioning step of 20 sinusoidal compression-tension cycles followed by 5 sinusoidal compression-tension cycles at different frequencies (0.1, 1, 5 Hz) using MTS Acumen 3 Electrodynamic Test System (MTS Systems Corporation, Eden Prairie, Minnesota, USA) equipped with a 500N load cell, also from MTS. Compression and tension loads were oscillated by set their values at -60 N and 45 N, respectively, with 60 N corresponding to one body weight (taking into consideration the average area of the IVD used in this study). The last cycle at each frequency was used to determine the biomechanical parameters of the specimen studied for the different groups considered in this work. It is important to say that Johannessen et al.5 required 20 cycles to observe constant cycle for the cyclic loading, but in our hand all 5 cycles were exactly similar (Figure S2, Supporting Information). Cyclic loading was followed by a creep test consisting of a compressive load of -30 N for 1 h.

11 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Neutral zone (NZ) length, range of motion (ROM), compressive stiffness (Scomp) and tensile stiffness (Stens) depicted in (Figure S3, Supporting Information) were determined.5 ROM was measured as the total peak-to-peak displacement. Compressive (Scomp) and tensile stiffness (Stens) were calculated by linear regression of the load-displacement data between the 70% and 100% of maximum load. To calculate NZ length, a third-order polynomial was fit to the compression-tension load-displacement data and its derivative was calculated. The displacement corresponding to the minimum of the derivative was used as inflection point between compression and tension. A line with a slope equal to the derivative of the polynomial model (as definition of NZ stiffness) was extended through the inflection point. The axial displacement corresponding with compressive and tension intercepts was taken as the NZ length.

High-resolution computed tomography (µCT) analysis High-resolution peripheral quantitative tomography (XtremeCT, Scanco Medical, Brüttisellen, Switzerland) was used to image each motion segment before and after mechanical test. All parameters for scanning were kept constant with a voltage of 60 kVp, integration time of 200 ms and nominal isotropic voxel size of 82 µm, with an equal in-plane and between-plane voxel size. The two-dimensional µCT images were used as standard convolution-back projection procedure. After pre-processing the images were segmented using the same values of thresholding and Gaussian filtering to extract the mineralized phase of the spine segment and the BaSO4 present in the hydrogels. Location and volume of the injected materials were assessed as well as deformation/displacement before and after mechanical test. A high-resolution X-ray computed tomography system (µCT 40, Scanco Medical, Brüttisellen, Switzerland) was used to image hydrogels (energy of 45 kV and an intensity of 176 µA) at a resolution of 7 µm voxel. To determine macroscopic damage of the 12 ACS Paragon Plus Environment

Page 12 of 35

Page 13 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

hydrogels after injection, the same sample procedure as the one described for the rheology was followed, but the hydrogels were analyzed by high-resolution X-ray tomography system described above.

Statistical analyses Kolmogorov-Smirnov normality tests were used to confirm the normal distribution of the data. For multiple group comparisons, we performed one-way ANOVA analysis using Bonferroni post-hoc analysis. Number of biological replicates (N) for each experiment and average ± SD are reported. A p ≤ 0.05 was considered statistically significant.

RESULTS AND DISCUSSIONS

Dynamic Properties of DH Hydrogels DH at 30 wt% exhibited self-healing properties as attested by its ability to readily regain its macroscopic integrity after damage (Figure 1B). The mechanical properties of DH and the corresponding CH prepared in phosphate buffer saline (PBS) solutions were studied by dynamic rheology at 37°C, which depicts shear elastic modulus (G') and shear loss modulus (G'') at a fixed strain of 1% (Figure 1, C-D). It is worth mentioning that the hydrogel state was confirmed for both materials with G' greater than G'' at all frequencies. DH exhibited frequency-dependent stiffness with a clear increase of G' with increasing frequency, from 4.5 kPa to 78 kPa. On the other hand, CH at the same (PEGSH)4 concentration exhibited constant values of G' and G'' over the whole frequency range at around 20 kPa and 100 Pa, respectively, which is characteristic of covalently cross-linked hydrogels. The increase of G' for DH resulted in a stiffer hydrogel compared to CH at frequencies higher than 2 Hz. This frequency-dependent mechanical property at physiological pH was attributed to Au-S/SS permanent exchange.37,38,39At low deformation frequencies, the exchange reaction was 13 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

sufficiently fast for the polymer network to rearrange despite deformation. However, at frequencies faster than the Au-S/SS exchange reaction the network was “frozen” with an increasing amount of polymer chains entanglements which leads to a stiffening of the hydrogel. Interestingly, it has been reported that vibrations in this range of 4-10 Hz and deformation below 1 % induce degenerative changes in the IVD.43 Thanks to the stiffening effect observed at increasing frequencies, DH hydrogel has the potential to counteract such IVD degeneration associated with high frequency vibrations. One important aspect of the current study concerned the localization of the hydrogels inside the IVD following cyclic loading. In order to visualize the hydrogels, BaSO4 powder was added as contrast agent for µCT. A slight effect on the rheological properties was observed in the presence of 20 wt% BaSO4 in the hydrogels (Figure 1 C-D). However, only minimal rheological changes were observed (< 10 %), so it is unlikely that the contrast agent disrupted the network in either hydrogels. Preliminary compression tests were carried out at different speeds using the same rheometer apparatus at fixed strain deformation to determine the resulting normal force, or resistance, offered by both hydrogels with BaSO4 (Figure 1, E-F). A protocol of 3 compression rates, namely 10, 100 and 1000 µm·sec-1, at 10 % deformation were selected. Faster compression speed rates resulted in an increase of the corresponding normal force (N) for DH, whereas CH showed similar response independently of the compression speed applied. This result confirmed the strain-rate dependent stiffness of DH under axial compression, which might indicate shock absorbing properties, in addition to frequency-dependent stiffness observed under oscillatory deformation (Figure 1, C-D). In addition, cell viability and proliferation using Cell Titter Glo and Live/Dead assay on days 1, 3 and 7 have been carried out to study the cytocompatibility of DH (Figure S4) and compared to cell alone (positive control) and in the presence of SDS (negative control). To do so, healthy bovine nucleus pulposus (NP) cells have been used as model cell to fit nucleus 14 ACS Paragon Plus Environment

Page 14 of 35

Page 15 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

pulposus application proposed in this study. NP cells viability and proliferation is clearly affected by DH with lower amount of cells compared to non-treated cultures. However, the luminescence values increased gradually from day 1 to 7 (Figure S4), indicating cells proliferation in the presence of the hydrogels. This result was confirmed by Live/Dead staining, after 7 days, with most of the cells appearing alive (green fluorescence) with no obvious difference in cell morphology while dead cells (red fluorescence) were hardly observed. Shock-absorbing Performance of DH in Nucleotomized Intervertebral Discs The potential use of DH containing Au-S species as an efficient NP replacement for IVD repair was investigated in a bovine ex-vivo model by performing axial compression-tension cycling testing at different frequencies and creep experiments (Figure 2). Axial compressiontension tests were carried out on 4 groups: (i) Intact IVD (ii) NP-nucleotomized IVD (iii) DH injected in NP-nucleotomized IVD and (iv) CH injected in NP-nucleotomized IVD. Force vs. Displacement curves were obtained at different frequencies, i.e. 0.1, 1.0 and 5.0 Hz, which correspond to normal physical activities, from normal walking to running (Figures 3A).44

15 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. (A) Representative Force vs. Displacement curves obtained during cyclic compression/tension load between -60 N (compression) and 45 N (tension) applied at 0.1, 1 and 5 Hz for one representative specimen of each group: (black) intact IVD, (red) NPnucleotomized IVD, (blue) DH injected NP and (green) CH injected NP. On the right handside is shown a higher magnification of the same Force vs. Displacement curve at small loads (between -2 N and 2 N) which corresponds to the neutral zone region where the role of NP is most significant. (B) Mean and standard deviation of the range of motion (ROM), neutral zone length and stiffness (NZ length and NZ stiffness, respectively) determined from force vs. displacement curve obtained during the axial compression/tension mechanical testing (60N/45N) carried out at 0.1 Hz, 1 Hz and 5 Hz for the specimen groups (N = 7) considered in this study: intact IVD (Intact), NP-nucleotomized IVD (NP-nucleotomized), DH injected into NP-nucleotomized IVD (DH) and CH injected into NP-nucleotomized IVD (CH). * indicates significant differences (p ≤ 0.05) while “ns” means that data did not show significant differences. At first glance, no dramatic differences were observed for the overall Force vs. Displacement curves of the intact IVD and DH injected IVD at the three frequencies studied, which might indicate the efficient replacement of NP tissue by DH. On the other hand, the similarity of the Force vs. Displacement curves for the nucleotomized IVD and CH injected IVD strongly suggests the lack of mechanical support conferred by the conventional hydrogel to the IVD. Statistical analysis of the 7 specimen of each groups confirmed that no significant differences were observed for ROM and NZ length for the intact IVD and DH injected IVD at all frequencies (Figure 3B). Thus, DH acts similarly to the native NP, cushioning the vertebras from the cyclic compression/tension stresses and recovering the similar 16 ACS Paragon Plus Environment

Page 16 of 35

Page 17 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

displacements as the intact IVD. Nucleotomized discs confirmed that the absence of NP significantly affected the mechanical performance of the overall IVD with an increase of ROM and NZ length of around 60 % and 150 %, respectively, compared to the intact IVD, independently of the frequency. These values are in good agreement with the ROM and NZ length obtained for radical nucleotomy5 and confirmed the crucial role of NP to dissipate energy, especially at weak load. The presence of CH was not able to reduce neither ROM nor NZ length, which demonstrates the poor contribution of the covalent hydrogel to restore the IVD mechanical integrity. Interestingly, an increase in NZ stiffness at higher frequencies was observed for the intact IVDs and NP-nucleotomized IVDs containing DH. Therefore, DH could also restore the frequency-dependent stiffness (or shock absorbing properties) of the NP tissue. The lack of significant differences observed between the group of intact IVD and the one with DH confirmed that our dynamic hydrogel could mimic the biomechanical role of healthy NP. It is worth mentioning that somewhat greater NZ stiffness values were measured using DH as NP replacement compared to the intact IVD at all frequencies, which confirmed again the use of DH as NP replacement. Nucleotomized IVD without hydrogel showed a decrease of NZ stiffness that was greater than 80 % compared to the one of intact IVDs. CH contribution to the NZ stiffness restoration of nucleotomized IVDs was only minimal. To the best of our knowledge, these results represent the first example of the application of a self-healing dynamic hydrogel inside IVD that exhibits similar shock-absorbing properties to the native NP tissue. Hence, this DH can be considered as a viable alternative to substitute damaged NP for IVD repair. It is noteworthy that a slight increase of NZ stiffness with increasing frequency was also observed in the absence of NP which has been attributed to remaining NP tissue (partial nucleotomy) in a previous study.5 However, the increase in NZ stiffness does not appear significant in the present work, probably due to the fact that the 17 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

nucleotomy was quite radical, thus leaving only a small amount of NP tissue. Conversely, this stiffening effect was completely lost when CH was used as NP replacement, which might indicate that the absence of dynamic properties does not allow the frequency dependent stiffness of NP to be recovered. It is also worth mentioning that compressive and tensile stiffness (Scomp and Sten, respectively), also extracted from Force vs. Displacement data, were similar for all groups, as those parameters are strongly dependent on the AF, but not the NP (Figure S5, Supporting Information).45

Figure 4. (Top left) Creep response to a compression load of -30 N applied to the IVD for 60 min. (Top left) Characteristic Displacement vs. Time curve for one specimen of each group: intact IVD (black), NP-nucleotomized IVD (red), DH injected in NP-nucleotomized IVD (blue) and CH injected in NP-nucleotomized IVD (green). (Top right) Representation of the average displacement at early creep stages (< 2 s) of the compression load applied (-30 N) for

18 ACS Paragon Plus Environment

Page 18 of 35

Page 19 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

one specimen of each group studied here. (Bottom) Table summarizing the total creep displacement and rate of deformation at early creep (< 2 s) for all groups (N = 7). Creep Behavior of Nucleotomized IVDs Supplemented with DH Hydrogels Creep studies were carried out for the different groups to study the role of NP when a constant load at -30N was applied (Figure 4). All samples displayed a similar response: i) an initial fast deformation during the early creep response (< 1 s) to the applied load and ii) a much slower deformation leading to an apparent plateau or equilibrium for late creep response (> 1 s). The displacement was much more important for NP-nucleotomized IVD compared to intact IVD which confirmed the important role of NP to dissipate the total load energy. The presence of DH allowed the creep displacement to be reduced. As shown in Figure 4, partial recovery of the viscoelastic properties of intact IVDs could be achieved. On the other hand, CH appeared to have no effect on the final deformation after 60 min load. This result confirmed the previous findings obtained for the cyclic compression-tension tests, i.e. the dynamic properties of DH are crucial to mimic the mechanical role of NP in the IVD and that CH has no effect on the overall biomechanical properties of the IVD. Interestingly, the main differences between the samples occurred at early creep response of the IVD which can be correlated to the ability of NP to dissipate the load through the whole IVD.6 The viscoelastic response at early creep, which can be considered proportional to the slope of the creep curves, was much faster in the absence of NP with a rate of displacement at around 3.23±0.75 mm s15

. Importantly, the presence of DH could slow down significantly the rate of displacement at

around 1.79±0.56 mm s-1 which is similar to the one obtained for the intact IVD (1.77±0.3 mm s-1). This behavior is characteristic of the shock absorbing properties of NP that confirms DH plays a similar role.6 On the other hand, CH showed the highest rate of deformation under creep (3.63±1.67 mm s-1), confirming its minimal contribution to the biomechanical integrity of the whole IVD. 19 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

In summary, the biomechanical property characteristics of healthy IVD could be recovered using DH due to its mechanical similarity to the native NP, especially its shock absorbing properties. However, the dramatic differences observed between IVD treated with either DH or CH were more surprising, especially that IVD treated with CH did not show any mechanical improvement and behaved like the nucleotomized IVDs. The absence of Au-S species for CH did not provide self-healing and frequency-dependent properties which could lead to inefficient filling of the nucleotomized space due to irreversible damage occurring during the injection process.

20 ACS Paragon Plus Environment

Page 20 of 35

Page 21 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Biomacromolecules

21

ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48

Page 22 of 35

Figure 5. (Top, black and white images) Representative µCT images and (Top, colored images) reconstructions of the IVD containing DH or CH before and after the entire mechanical testing, i.e. compression/tension at different frequency and creep tests. 3D reconstruction images were performed using transversal-sectional imaging (through the z axis) with XtremCT of the IVD containing DH and CH before and after mechanical testing. Note that both hydrogels were prepared with 20 wt% BaSO4 as contrast agent to allow its visualization with µCT (scale bar 5 mm for all images, EP: End plate, NP: Nucleus pulposus, AF: Annulus fibrosus, VB: Vertebral body; SIL: Silicone sealing). (Bottom) Table summarizing the total volume of the IVD and the volume-average of the hydrogels before and after mechanical testing. The percentage of remaining hydrogel is also indicated.

22

ACS Paragon Plus Environment

Page 23 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Hydrogel Localization Before and After Biomechanical Testing µCT analysis was performed to visualize BaSO4 dispersed in the hydrogels and infer the location, potential displacement of the hydrogels from their original position and damage after mechanical testing (Figure 5). The cyclic compression-tension tests followed by the creep experiments did not damage the injected DH material at the observed tomography resolution of 82 µm. The absence of leakage or diffusion of DH in the surrounding AF was also confirmed by the volume of hydrogels, which remained unchanged after biomechanical testing (around 97% of its original volume) (Table in Figure 5). On the other hand, the volume of CH after mechanical testing decreased significantly to around 80% of its original volume. In both hydrogels, the contrast agent was not chemically bound to the (PEGSH)4 network. Therefore, it cannot be excluded that the contrast agent remained into the NP cavity while some polymer chains diffused into the surrounding tissue. Visual inspection of the transversal cuts of the IVDs specimen containing the hydrogels after mechanical testing showed that the NP cavity remained filled but the native tissue and the injected material were not clearly distinguishable (Figure S6 in Supporting Information). On the other hand, examination of the specimens from the negative control group, i.e. NP-nucleotomized IVD, by µCT confirmed that the cavity, albeit reduced, was still present after the mechanical tests (Figure S7 in Supporting Information). Hence, if the diffusion of polymeric material into the surrounding tissue occurred it is likely to be in relatively small quantities due to the similar shape of the overall implant and the presence of the materials after mechanical testing, as judged by µCT imaging and transversal cut observation, respectively. Self-Healing Properties of Injected DH As previously mentioned, the hydrogels were still present inside the IVD after mechanical testing, but it proved to be very difficult to distinguish from the surrounding IVD tissue. Therefore, the hydrogels could not be extracted from the NP cavity to study possible damages 23 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

suffered during specimen preparation and/or mechanical tests. In this context, continuous step strain experiments were carried out to determine the damage-healing property of both hydrogels (Figure 6). First, strain sweep experiments were carried out at 0.1 Hz to determine the strain required to damage the network of DH, i.e. G’ < G’’, which was found to be at strain greater than 700 %. Thus, continuous step strain consisting of successive strain at 1% and 800 % was applied to DH at 0.1 Hz. As expected, the dramatic increase of strain resulted in significant damage, with G’ < G’’. However, thanks to Au-S/SS exchange, the hydrogel state was recovered at 1 % strain with G’ reaching more than 90% of its original value. Further strain cycles showed that after further damage of the hydrogel microstructure at 800 % strain, similar G’ to the one obtained after the first damages were achieved, demonstrating the self-healing ability of DH. It is important to mention that in our case the original G’ could not be completely recovered after the first strain increase compared to self-healing hydrogels using others DCCs. 33,27 However, the relatively high G’ of our DH, at 10,000 Pa compared to 1,000 Pa for previously reported dynamic hydrogel, could explain the incomplete recovery in the time-scale of the experiments. On the other hand, similar experiments were carried out for CH. Obviously, CH appeared irreversibly damaged after applying 800% strain for the first time, as G’’ was greater than G’ after 1 cycle and the gel state could not be recovered. Therefore, the absence of Au-S/SS exchange confirmed that CH did not exhibit self-healing properties, as expected for conventional covalent hydrogel. It can also be anticipated that injection of CH through a syringe will result in further damage to the covalent network.

24 ACS Paragon Plus Environment

Page 24 of 35

Page 25 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

Figure 6. Damage-healing property of (left) DH and (right) CH demonstrated by continuous step strain (1% strain→800% strain→1% strain) measurements at 0.1 Hz and 37 °C depicting the variation of G’ (■) and G’’ (●). Additional rheological studies and µCT imaging of the preformed hydrogels before and after injection into a disc mold were carried out to mimic the injection of the hydrogels in the cavity of the NP-nucleotomized IVD (Figure 7). Frequency sweep studies of CH showed a dramatic decrease of G', from 22 kPa before injection to 5 kPa after injection (Figure 7B), which was likely due to irreversible damages suffered by the polymer network. It is important to keep in mind that CH in the absence of Au-S species did not exhibit the self-healing property, as demonstrated in Figure 6. µCT images performed at a resolution of 7 µm voxel, demonstrated that CH not only suffered microscopic, but also macroscopic damages following injection (Figure 7D and 7F). Therefore, the very poor mechanical properties of CH can be attributed to the damage suffered during its injection, before mechanical testing. In contrast, both G' and G'' for DH remained unchanged after injection, as judged by rheological studies (Figure 6A). Thanks to the presence of Au-S species which conferred self-healing properties, DH recovered its mechanical properties after injection and resting overnight in a confined space. µCT confirmed the preservation of the DH mechanical integrity following injection (Figure 7C and 7E), thus confirming the behavior observed with the restored IVDs. 25 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Therefore, in addition to the frequency-dependent stiffness properties that mimic the mechanical properties of healthy NP, the self-healing ability of DH conferred by Au-S/SS exchange reaction suggests that this material could have a long-life span when used as injectable NP substitute.

Figure 7. Frequency sweep studies depicting G’ (filled symbols) and G’’ (opened symbols) for (A) DH and (B) (blue) CH before and (red) after injection with a 2 mL syringe. µCT images of DH (C) before and (E) after injection and CH (D) before and (F) (scale bar = 1 mm). For all experiments, the hydrogels were injected in a cylinder mold and let self-healed overnight with slight pressure to mimic the conditions after injection in the cavity of nucleotomized IVD. In this study we have demonstrated that preformed DH hydrogel are capable to self-heal following injection and restore the mechanical properties of IVDs to a level close to intact IVDs, including frequency-dependent stiffness. Preformed DH hydrogels may represent a

26 ACS Paragon Plus Environment

Page 26 of 35

Page 27 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

safer option compared to in-situ forming hydrogels, as the latter may experience diffusion of precursor solutions in the surrounding tissue, resulting in reproducibility issues. CONCLUSIONS In this work, we have demonstrated that preformed DH containing Au-S species could be used as an efficient injectable NP replacement to treat IVD repair. This is the first application of self-healing dynamic hydrogels inside biological tissues, namely IVD. The dynamic character of the hydrogel due to Au-S/SS exchange conferred self-healing properties and frequency-dependent stiffness (like a shock absorber). The biomechanical parameters (ROM, NZ length and stiffness) of nucleotomized IVDs treated with preformed DH hydrogel were fully recovered, achieving similar levels to intact IVDs. Moreover, an increasing NZ stiffness at increasing frequencies could be restored for IVD treated and creep experiments confirmed that DH could dissipate the energy of the load similarly to healthy NP. On the other hand, conventional CH did not have any beneficial impact on the overall IVD mechanical properties with similar results as the one obtained for NP-nucleotomized IVD, due to the irreversible damages suffered by CH after injection, as shown by frequency sweep and continuous step strain studies. Thus, in addition to exhibit shock absorber properties to mimic the role of healthy NP at increasing frequencies, self-healing of DH is also crucial to recover the material mechanical integrity after injection. Finally, µCT confirmed that DH was not dislocated following cyclic loading and therefore indicate for the first time the potential use of selfhealing dynamic hydrogels for treatment of IVD degeneration.

Supporting Information. Representative cyclic load displacement data, Representative compression-tension curve with corresponding curve fits, Tensile and compressive stiffness, µCT analysis of NP-nucleotomized IVD before and after mechanical testing and box charts

27 ACS Paragon Plus Environment

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

representing all samples for each group used for mechanical testing. The Supporting Information is available free of charge on the ACS Publications website. Corresponding Author *Dr Damien Dupin ([email protected]), Dr Iraida Loinaz ([email protected]). Author Contributions The manuscript was written through contributions of all authors. All authors have given approval to the final version of the manuscript. ‡These authors contributed equally. Funding Sources Dr. A. Pérez-San Vicente acknowledges the financial support of COST Action STSM (COSTSTSM-MP1005-17010 and COST-STSM-MP1005-14449) and SNF international short visit grant (IZKOZ3-161639). ACKNOWLEDGMENT The authors would like to gratefully acknowledge Dieter Wahl for excellent support in the development of the biomechanical test setup and Silvia Pettinelli for assistance in motion segment preparation.

28 ACS Paragon Plus Environment

Page 28 of 35

Page 29 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

References (1) Dagenais, S.; Caro, J.; Haldeman, S., A systematic review of low back pain cost of illness studies in the United States and internationally. Spine J 2008, 8 (1), 8-20. (2) Chan, S. C.; Gantenbein-Ritter, B., Intervertebral disc regeneration or repair with biomaterials and stem cell therapy--feasible or fiction? Swiss Med Wkly 2012, 142, w13598. (3) O'Neill, C. W.; Kurgansky, M. E.; Derby, R.; Ryan, D. P., Disc stimulation and patterns of referred pain. Spine (Phila Pa 1976) 2002, 27 (24), 2776-81. (4) Balague, F.; Mannion, A. F.; Pellise, F.; Cedraschi, C., Non-specific low back pain. Lancet 2012, 379 (9814), 482-91. (5) Johannessen, W.; Cloyd, J. M.; O'Connell, G. D.; Vresilovic, E. J.; Elliott, D. M., Transendplate nucleotomy increases deformation and creep response in axial loading. Ann Biomed Eng 2006, 34 (4), 687-96. (6) Zhou, Z.; Gao, M.; Wei, F.; Liang, J.; Deng, W.; Dai, X.; Zhou, G.; Zou, X., Shock absorbing function study on denucleated intervertebral disc with or without hydrogel injection through static and dynamic biomechanical tests in vitro. Biomed Res Int 2014, 2014, 461724. (7) Guterl, C. C.; See, E. Y.; Blanquer, S. B.; Pandit, A.; Ferguson, S. J.; Benneker, L. M.; Grijpma, D. W.; Sakai, D.; Eglin, D.; Alini, M.; Iatridis, J. C.; Grad, S., Challenges and strategies in the repair of ruptured annulus fibrosus. Eur Cell Mater 2013, 25, 1-21. (8) Takahashi, K.; Aoki, Y.; Ohtori, S., Resolving discogenic pain. Eur Spine J 2008, 17 Suppl 4, 428-31. (9) Anderson, D. G.; Risbud, M. V.; Shapiro, I. M.; Vaccaro, A. R.; Albert, T. J., Cell-based therapy for disc repair. Spine J 2005, 5 (6 Suppl), 297S-303S.

ACS Paragon Plus Environment

29

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 30 of 35

(10) Vernengo, J.; Fussell, G. W.; Smith, N. G.; Lowman, A. M., Synthesis and characterization of injectable bioadhesive hydrogels for nucleus pulposus replacement and repair of the damaged intervertebral disc. Journal of Biomedical Materials Research Part B: Applied Biomaterials 2010, 93B (2), 309-317. (11) Roughley, P. J., Biology of intervertebral disc aging and degeneration: involvement of the extracellular matrix. Spine 2004, 29 (23), 2691-2699. (12) Grad, S., Alini, M., Eglin, D., Sakai, D., Mochida, J., Mahor, & Pandit, A., Cells and biomaterials for intervertebral disc regeneration. In Synthesis lectures on tissue engineering, 2010; Vol. 5(1), 1-104 (13) Milani, A. H.; Freemont, A. J.; Hoyland, J. A.; Adlam, D. J.; Saunders, B. R., Injectable doubly cross-linked microgels for improving the mechanical properties of degenerated intervertebral discs. Biomacromolecules 2012, 13 (9), 2793-801. (14) Wojtecki, R. J.; Meador, M. A.; Rowan, S. J., Using the dynamic bond to access macroscopically responsive structurally dynamic polymers. Nat Mater 2011, 10 (1), 14-27. (15) Kloxin, C. J.; Bowman, C. N., Covalent adaptable networks: smart, reconfigurable and responsive network systems. Chemical Society Reviews 2013, 42 (17), 7161-7173. (16) Kloxin, C. J.; Scott, T. F.; Adzima, B. J.; Bowman, C. N., Covalent Adaptable Networks (CANs): A Unique Paradigm in Crosslinked Polymers. Macromolecules 2010, 43 (6), 26432653. (17) García, F.; Smulders, M. M., Dynamic covalent polymers. Journal of Polymer Science Part A: Polymer Chemistry 2016, 54 (22), 3551-3577.

ACS Paragon Plus Environment

30

Page 31 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

(18) Cromwell, O. R.; Chung, J.; Guan, Z., Malleable and Self-Healing Covalent Polymer Networks through Tunable Dynamic Boronic Ester Bonds. J Am Chem Soc 2015, 137 (20), 6492-5. (19) Taylor, D. L., Self‐Healing Hydrogels. Advanced Materials 2016. (20) Roberts, M. C.; Hanson, M. C.; Massey, A. P.; Karren, E. A.; Kiser, P. F., Dynamically restructuring hydrogel networks formed with reversible covalent crosslinks. Advanced Materials 2007, 19 (18), 2503-2507. (21) He, L.; Fullenkamp, D. E.; Rivera, J. G.; Messersmith, P. B., pH responsive self-healing hydrogels formed by boronate–catechol complexation. Chemical Communications 2011, 47 (26), 7497-7499. (22) Skardal, A.; Zhang, J.; McCoard, L.; Oottamasathien, S.; Prestwich, G. D., Dynamically crosslinked gold nanoparticle - hyaluronan hydrogels. Adv Mater 2010, 22 (42), 4736-40. (23) Deng, G.; Li, F.; Yu, H.; Liu, F.; Liu, C.; Sun, W.; Jiang, H.; Chen, Y., Dynamic hydrogels with an environmental adaptive self-healing ability and dual responsive sol–gel transitions. ACS Macro Letters 2012, 1 (2), 275-279. (24) Amamoto, Y.; Kamada, J.; Otsuka, H.; Takahara, A.; Matyjaszewski, K., Repeatable photoinduced self-healing of covalently cross-linked polymers through reshuffling of trithiocarbonate units. Angew Chem Int Ed Engl 2011, 50 (7), 1660-3. (25) Fairbanks, B. D.; Singh, S. P.; Bowman, C. N.; Anseth, K. S., Photodegradable, photoadaptable hydrogels via radical-mediated disulfide fragmentation reaction. Macromolecules 2011, 44 (8), 2444-2450.

ACS Paragon Plus Environment

31

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 32 of 35

(26) Zhang, Y.; Tao, L.; Li, S.; Wei, Y., Synthesis of multiresponsive and dynamic chitosanbased hydrogels for controlled release of bioactive molecules. Biomacromolecules 2011, 12 (8), 2894-2901. (27) Tseng, T. C.; Tao, L.; Hsieh, F. Y.; Wei, Y.; Chiu, I. M.; Hsu, S. H., An Injectable, SelfHealing Hydrogel to Repair the Central Nervous System. Adv Mater 2015, 27 (23), 3518-24. (28) Yang, B.; Zhang, Y.; Zhang, X.; Tao, L.; Li, S.; Wei, Y., Facilely prepared inexpensive and biocompatible self-healing hydrogel: a new injectable cell therapy carrier. Polymer Chemistry 2012, 3 (12), 3235-3238. (29) Ding, F.; Wu, S.; Wang, S.; Xiong, Y.; Li, Y.; Li, B.; Deng, H.; Du, Y.; Xiao, L.; Shi, X., A dynamic and self-crosslinked polysaccharide hydrogel with autonomous self-healing ability. Soft Matter 2015, 11 (20), 3971-3976. (30) Lü, S.; Gao, C.; Xu, X.; Bai, X.; Duan, H.; Gao, N.; Feng, C.; Xiong, Y.; Liu, M., Injectable and self-healing carbohydrate-based hydrogel for cell encapsulation. ACS applied materials & interfaces 2015, 7 (23), 13029-13037. (31) Deng, C. C.; Brooks, W. L.; Abboud, K. A.; Sumerlin, B. S., Boronic acid-based hydrogels undergo self-healing at neutral and acidic pH. ACS Macro Letters 2015, 4 (2), 220-224. (32) Cash, J. J.; Kubo, T.; Bapat, A. P.; Sumerlin, B. S., Room-temperature self-healing polymers based on dynamic-covalent boronic esters. Macromolecules 2015, 48 (7), 2098-2106. (33) Barcan, G. A.; Zhang, X.; Waymouth, R. M., Structurally dynamic hydrogels derived from 1,2-dithiolanes. J Am Chem Soc 2015, 137 (17), 5650-3. (34) Zhang, X.; Waymouth, R. M., 1,2-Dithiolane-Derived Dynamic, Covalent Materials: Cooperative Self-Assembly and Reversible Cross-Linking. Journal of the American Chemical Society 2017 139, 3822-3833.

ACS Paragon Plus Environment

32

Page 33 of 35

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Biomacromolecules

(35) Rodell, C. B.; Kaminski, A. L.; Burdick, J. A., Rational design of network properties in guest–host assembled and shear-thinning hyaluronic acid hydrogels. Biomacromolecules 2013, 14 (11), 4125-4134. (36) Boekhoven, J.; Rubert Pérez, C. M.; Sur, S.; Worthy, A.; Stupp, S. I., Dynamic display of bioactivity through host–guest chemistry. Angewandte Chemie International Edition 2013, 52 (46), 12077-12080. (37) Casuso, P.; Perez-San Vicente, A.; Iribar, H.; Gutierrez-Rivera, A.; Izeta, A.; Loinaz, I.; Cabanero, G.; Grande, H. J.; Odriozola, I.; Dupin, D., Aurophilically cross-linked "dynamic" hydrogels mimicking healthy synovial fluid properties. Chem Commun (Camb) 2014, 50 (96), 15199-201. (38) Casuso, P.; Odriozola, I.; Perez-San Vicente, A.; Loinaz, I.; Cabanero, G.; Grande, H. J.; Dupin, D., Injectable and Self-Healing Dynamic Hydrogels Based on Metal(I)-Thiolate/Disulfide Exchange as Biomaterials with Tunable Mechanical Properties. Biomacromolecules 2015, 16 (11), 3552-61. (39) Gantar, A., Drnovšek, N., Casuso, P., Pérez-San Vicente, A., Rodriguez, J., Dupin, D., ... & Loinaz, I., Injectable and self-healing dynamic hydrogel containing bioactive glass nanoparticles as a potential biomaterial for bone regeneration. RSC Advances 2016, 6(73), 69156-69166. (40) Vadala, G.; Russo, F.; Pattappa, G.; Schiuma, D.; Peroglio, M.; Benneker, L. M.; Grad, S.; Alini, M.; Denaro, V., The transpedicular approach as an alternative route for intervertebral disc regeneration. Spine (Phila Pa 1976) 38 (6), E319-24. (41) Boxberger, J. I.; Orlansky, A. S.; Sen, S.; Elliott, D. M., Reduced nucleus pulposus glycosaminoglycan content alters intervertebral disc dynamic viscoelastic mechanics. J Biomech 2009, 42 (12), 1941-6.

ACS Paragon Plus Environment

33

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 34 of 35

(42) Barbir, A.; Michalek, A. J.; Abbott, R. D.; Iatridis, J. C., Effects of enzymatic digestion on compressive properties of rat intervertebral discs. J Biomech 2010, 43 (6), 1067-73. (43) Pope, M. H.; Magnusson, M.; Wilder, D. G., Kappa Delta Award. Low back pain and whole body vibration. Clin Orthop Relat Res 1998, (354), 241-8. (44) Fakhari, A.; Phan, Q.; Thakkar, S. V.; Middaugh, C. R.; Berkland, C., Hyaluronic acid nanoparticles titrate the viscoelastic properties of viscosupplements. Langmuir 2013, 29 (17), 5123-31. (45) Shirazi-Adl, A., Finite element stress analysis of a push-out test part 1: fixed interface using stress compatible elements. Journal of biomechanical engineering 1992, 114 (1), 111-118.

“Self-Healing Dynamic Hydrogel as Injectable Shock-Absorbing Artificial Nucleus Pulposus” by Adrián Pérez-San Vicente, Marianna Peroglio, Manuela Ernst, Pablo Casuso, Iraida Loinaz,* Hans-Jürgen Grande, Mauro Alini, David Eglin, Damien Dupin* TOC

ACS Paragon Plus Environment

34

Page 35 of 35

Biomacromolecules

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 ACS Paragon Plus Environment

35