Affibody-Functionalized Hyaluronic Acid Hydrogels for Bone Regeneration

Introduction: Bone morphogenetic protein 2 (BMP-2) has shown promise in promoting bone formation; however, its uncontrolled delivery can lead to adverse effects such as inflammation and ectopic bone formation, necessitating the use of a suitable biomaterial carrier to achieve controlled release. Clinically, collagen sponges are used as natural carriers that have a weak affinity for BMP-2 but do not provide sustained release. One of the major drawbacks of this system is the diffusion-controlled release which results in a burst release of BMP-2 which can lead to complications including local inflammation and ectopic bone formation. We sought to use hyaluronic acid hydrogels because they have low immunogenicity, are biodegradable, and chemically modifiable for hydrogel formation and peptide conjugation. To enable the sustained release of BMP-2, we employed the use of affibodies, which are small proteins that can bind to specific target proteins with different affinities. Our lab has discovered an affibody that is specific to BMP-2 and demonstrated sustained BMP-2 release using this affibody in a 4arm poly(Ethylene Glycol)-Maleimide (PEG4Mal) hydrogel system.[1] Our objective was to engineer a hyaluronic acid hydrogel functionalized with affibodies to control the release of BMP-2 and form bone in a subcutaneous/ectopic bone mineralization model. We hypothesized that HA hydrogels within collagen sponges would increase bone formation in-vivo. We also hypothesized that affibody-conjugated HA hydrogels would increase the encapsulation of BMP-2 while slowing down its release.

Materials and

Methods: As previously described[2], sodium hyaluronate (HA,100kDa) was modified with either adipic acid dihydrazide forming HA-ADH, or exposed to sodium periodate for oxidation to reveal aldehydes (HA-OX). The hydrazines on HA-ADH are then able to react with the aldehydes on HA-OX and form a hydrazone-crosslinked HA hydrogel. HA-ADH was further modified with a norbornene functional group where affibodies could be conjugated through a photo-initiated reaction using Irgacure2959. Affibody conjugation was assessed using Pierce-660 protein reagent. Release of BMP-2 was assessed from HA hydrogels with and without affibodies, collagen sponge alone, and collagen sponge with HA with and without affibodies. For in vitro protein release and subcutaneous mineralization studies, HA hydrogels were polymerized within 6-8mm diameter Medtronic collagen sponges. 50µL of HA-ADH or HA-ADH-Affibody was first dropped onto the sponge, then 50µL of HA-OX containing 2ng/µL of BMP-2 was added and hydrogels were allowed to crosslink for 30 minutes. Collagen sponges without HA hydrogels were rehydrated with 100µL of 1ng/µL BMP-2. Constructs were incubated in 0.1% w/v BSA in PBS at 37°C, and the supernatant was sampled over 28 days. BMP-2 in the supernatant was quantified through an ELISA.
For assessing mineralization in vivo, collagen sponges or collagen sponges with HA hydrogels were synthesized with 0,1.0,2.5, or 5.0 µg of BMP-2 (n=4 per dose/construct). Four pockets were made on the dorsal region of female Sprague Dawley rats, and constructs were implanted subcutaneously. After four weeks, bone formation and scaffold degradation were evaluated using micro-computed tomography (µCT), x-ray imaging, and histology.

Results, Conclusions, and Discussions: HA hydrogels with and without affibodies inside a collagen sponge had nearly 100% loading efficiency (Figure 1.A). The release study showed that over 48 hours, the HA and CS alone groups released the most BMP-2 (Figure 1.B). Collagen sponges with either HA or affibody-conjugated HA released less than 20% of the protein.
µCT analysis after four weeks showed that there was no subcutaneous mineralization in either CS or CS+HA group at 0 or 1µg BMP-2. At 2.5µg BMP-2, CS and CS+HA showed comparable bone volume with the mean being 0.3382 ± 0.6502 and 0.6840 ± 0.8884 mm3 for the CS and CS+HA respectively. At 5µg BMP-2, CS+HA constructs had significantly more bone volume (6.488 ± 2.387) compared to CS alone (0.9915 ± 0.8881).
Our in vitro study demonstrates how conjugating affibodies onto HA increases BMP-2 retention and slows release. The amount of BMP-2 encapsulated inside collagen sponges increased from 91.91% to 99.13% and 99.54% when HA or HA-Affibody respectively, was polymerized in-situ. At 48 hours, collagen sponges without hydrogels had released 25.51% ± 4.63 of BMP-2, while those with HA and HA-Affibody only released 7.63% ± 1.63 and 4.26% ± 1.64 respectively. This lower amount suggests that BMP-2 is being retained within the constructs for longer.
The µCT analysis indicated that lower doses of BMP-2 are insufficient to induce bone formation, highlighting the necessity of an optimal BMP-2 dosage for effective bone regeneration. At a BMP-2 dose of 2.5 µg, both CS and CS+HA groups showed comparable bone volumes, and the presence of the HA hydrogel did not significantly enhance bone volume compared to the CS alone. However, at a BMP-2 dose of 5 µg, a significant difference emerged between groups, suggesting that the hydrogel mitigates the initial burst release typically associated with collagen sponges, providing a more consistent and prolonged release of BMP-2.
Future studies will focus on assessing BMP-2 bone formation from affibody-conjugated hydrogels subcutaneously while also continuing in vitro BMP-2 release at least 21 days. We will also assess the bone regenerative capability of these hydrogels in a critically-sized femoral bone defect model.

Acknowledgements (Optional):

References:
[1] Dorogin, J. et al. Moderate-Affinity Affibodies Modulate the Delivery and Bioactivity of Bone Morphogenetic Protein-2. Advanced Healthcare Materials n/a, 2300793.
[2] Mozipo, E. A. et al. Statistical optimization of hydrazone-crosslinked hyaluronic acid hydrogels for protein delivery. J. Mater. Chem. B 12, 2523–2536 (2024).