Introduction: Hydrogels are playing an increasingly important role in the development of regenerative approaches for the intervertebral disc (IVD)^1,2. Since there is limited availability of native human discs for research and often animal models do not mimic human disc degeneration, testing the biomechanical performance of a hydrogel after implantation remains difficult. The objective of the present study was to adapt and optimize an in vitro model of bovine tail discs for biomechanical testing. Discs were artificially degenerated with different enzymes and the distribution of a hydrogel after injection was investigated.

Methods: In total, four groups of motion segments (CY3/4, n=6/group) were prepared from 24 fresh bovine tails and embedded in PMMA. Each group was injected with 0.25 or 5 U/mL chondroitinase ABC (ChABC), 65 U/mL papain or PBS (sham control), and cultured for 7 days (6% O₂, 37°C). Complex loading was applied to diminish the swelling of the discs after culture. As much as possible of a radiopaque albumin/hyaluronan hydrogel (Albugel) was injected and the injected volume quantified. µCT scans were performed after injection to view the material distribution within the discs. After each step, the biomechanical behavior of the motion segments namely range of motion (ROM) and disc height were determined. Statistics: Shapiro-Wilk, Mann-Whitney-U, Friedman, Bonferroni-post-hoc (p≤0.05).

Results: At day 7, all specimens digested with papain developed a cavity in the nucleus, in all other three groups the nucleus seemed macroscopically intact. After incubation, the ROM for papain remained unaltered in comparison to intact specimens (p=1.000), whereas PBS and ChABC groups became stiffer, i.e. ROM decreased (p≤0.250). Only following complex loading, the ROM significantly increased in all groups versus intact, with the most prominent increase for papain in axial rotation (p≤0.028). More hydrogel could be injected into the papain group (1.3±0.3mL) than into ChABC-5U-digested discs (0.7±0.25mL). µCT reconstruction of the IVDs showed one large bubble for papain and a more inhomogeneous fluffy-cloud-like distribution of the hydrogel in ChABC-5U-digested specimens (Fig. 1). The hydrogel decreased the ROM in both digested groups compared to the intact condition, more strongly in the ChABC-5U- than in the papain-digested group (Fig. 2, exemplary flexion-extension).

Discussion: The ROM increase and loss of disc height, as well as cavity formation in the papain-treated group indicate a similar behavior as described for human discs with cavities, e.g., due to disc herniation or nucleotomy³. This allows the targeted simulation of such pathologies. ChABC more closely mimics human degeneration occurring without cavities. The lower concentration of ChABC led to little effects and could mimic mild degeneration, whereas the higher concentration is suitable to simulate progressed degeneration. Both ChABC-5U and papain allowed standardized hydrogel injections and testing.However, differences in hydrogel distribution and injectable volume could be noticed. We hypothesize that the specific digestion of glycosaminoglycans by ChABC may lead to different structural defects than papain. These results have improved our overall understanding of the biomechanical effects (especially ROM) of IVD tissue digestion with ChABC and papain, and suggest that hydrogels can be investigated with these models. Acknowledgments: iPSpine (825925).