A self-healing radiopaque hyaluronic acid hydrogel as a new injectable biomaterial for precision medicine in osteoarthritis

Abstract

Rationale: Osteoarthritis (OA) is a degenerative disease affecting cartilage, synovium and bone, that is a major cause of pain and disability. Intra-articular injection of hyaluronic acid (HA) derivatives, also known as viscosupplementation (VS), is a common treatment for the symptomatic management of knee OA. Despite its widespread use, the magnitude of the clinical benefit of VS remains controversial, with conflicting results due to methodological differences and possible differences in efficacy between products related to remanence and rheological properties. Methods: Here, to create an effective HA-based treatment, an injectable self-healing HA hydrogel with long-persistent radiopacity is formed by tethering a clinical iodine contrast agent to HA. The labeling conditions are tuned to obtain sufficient X-ray signal without altering the biocompatibility, rheological and injectability properties of the hydrogel. Results: The iodine labeling enabled to monitor not only delivery of the hydrogel but also its retention in mouse knees up to 5 weeks post-administration using synchrotron K-edge subtraction-computed tomography. We further demonstrated that the unique properties of this hydrogel enable creation of a transient HA network in vivo that attenuates OA progression in a mouse model of OA. Moreover, our data showed that the rate of HA-I disappearance appears to predict treatment response, likely because a rapid elimination serves as an indirect indicator of in situ inflammation. Conclusion: Collectively, these results show that our radiopaque HA-I hydrogel holds significant promise for improving patient management in the treatment of OA before clinical symptoms worsen. Its capacity for in vivo tracking over time allows for personalized treatment schedules based on observed retention and therapeutic effect. As a result, this theranostic hydrogel emerges as a strong candidate for precision medicine in OA.

Keywords: Injectable hydrogel; Iodine; X-ray; hyaluronic acid; viscosupplementation.

Figure 1

Schematic illustration of the radiopaque and self-healing hyaluronic acid (HA) hydrogel for intra-articular injection in OA. The dynamic cross-links based on boronate ester bonds in the hydrogel network makes it injectable and capable of self-healing almost instantly. The iodine contrast agent (CA) labeling enables monitoring of hydrogel delivery and retention in the knee joint in a mouse model of OA up to 5 weeks post-administration using synchrotron K-edge subtraction computed tomography (SKES-CT). Therapeutic effects are evaluated post-mortem using biological analyses of cartilage and bone degradation.

Figure 2

Synthesis of the iodine-labeled HA gel precursors. A) Modification of hyaluronic acid 1 with an iodine-based contrast agent (AcTIB-NH₂ 2), affording HA-TIB 3. B) Grafting of either fructosamine 4 or 3-aminophenylboronic acid 6 on HA-TIB to obtain HA-TIB-Fru 5 and HA-TIB-PBA 7.

Figure 3

A) Frequency dependence of the storage modulus (G') and loss modulus (G'') of the HA-I hydrogel measured with 10% strain at 25° C and 37° C. B) Variation of G' and G'' when increasing strain values to 800% (hydrogel disruption), followed by reducing the strain to a constant value of 10% (linear viscoelastic region). C) Alternate step strain sweep tests with alternating strain deformations of 10 and 800% at a fixed frequency (1 Hz). D) Photo of hydrogel injection in an agarose phantom through a 26G (0.46 mm diameter) needle (neutral red was added to color the hydrogel for visualization only).

Figure 4

Imaging of the HA-I hydrogel in the knees of healthy mice with SKES-CT. Results for each knee are displayed on each row. A) Attenuation images (representative single slice from 3D data set). B) Corresponding iodine concentration maps. C) 3D view of segmented bone (white) and iodine (blue).

Figure 5

Imaging of the HA-I hydrogels with SKES-CT in the knees of OA mice. Results of 3 representative knees imaged at 3 different times post-administration are displayed on each row (24 h, n = 3; 48 h, n = 3; 72 h, n = 5). A) Attenuation images (representative single slice from 3D dataset). B) Corresponding iodine concentration maps. C) 3D view of segmented bone (white) and iodine (blue).

Figure 6

Animal experiment procedure to investigate the intra-articular location/retention of the hydrogel after injection and its therapeutic effect.

Figure 7

Imaging and quantification of the HA-I hydrogel with SKES-CT in the knees of OA mice (white for bone and blue for iodine). A) Three representative knees imaged on the day of injection (day 7, group HA-ICT). B) Quantification of the volume of HA-I hydrogel in knee joints at day 7 and day 42 in group HA-ICT (mean ± SEM). C) Images of the knee joint of a mouse taken at day 7 and day 42 (group HA-ICT). D) Three representative knees imaged at day 42 (group HA-I). E) Quantification of the hydrogel volume in knee joints at day 42 in group HA-I (mean ± SEM).

Figure 8

Protective effects of the HA-I hydrogel in the collagenase-induced osteoarthritis murine model. A) Representative 2D images of the lateral epiphysis of mice imaged post mortem by conventional µCT at day 42 (upper panel). Groups correspond to non-treated mice (NT) or mice injected with the iodinated HA gel with an additional SKES-CT in vivo imaging at day 7 (HA-ICT) or without (HA-I). Histomorphometric parameters of sub-chondral bone plates (lower panel; bone surface (BS), bone volume (BV)). B) Representative post mortem 3D conventional µCT images of the joints at day 42 showing ectopic calcifications of menisci and ligaments of the joint and quantification of calcified bone volumes. C) Osteoarthritis (OA) score and representative images of histological sections from the three groups of mice. Results are expressed as mean ± SEM. * p < 0.05; ** p < 0.01; *** p < 0.001 (Statistical analysis used the Mann-Whitney test (A,C: n = 30 including lateral and median plateaux; B: n = 15 entire joints).