Improving gene activated matrices for orthopaedic bone regeneration: alternative biomaterials and predicted mRNA structure based BMP2 signal peptide optimisation

Critical and non-union defects are bone injuries that will not fully heal spontaneously, with existing
treatments frequently unable to improve outcomes. These defects represent an increasing burden on
health systems world-wide and there is an unmet clinical need for improved treatments.
Gene-activated matrices (GAMs) combine a biomaterial and a gene therapy, attempting to improve
performance by providing multiple osteogenic stimuli simultaneously. This project aims to develop a
GAM from a self-assembling peptide (SAP) P11-4 with known hard tissue regeneration capabilities and
plasmid vectors encoding the osteogenic growth factor bone morphogenetic protein 2 (BMP2),
resulting in an injectable GAM that promotes a multifaceted approach to osteogenesis.
New reporter vectors were created and transfection optimisation experiments performed to facilitate
the P11-4 GAM work. Transfection from P11-4 GAMs proved difficult and several experiments
investigating vector polyplex size and hydrogel porosity were performed to elucidate the mechanisms.
A cross-linked gelatine cryogel was then developed as a passive scaffold to allow continued
assessment of the plasmid vectors, demonstrating suitable porosity, sterility and biocompatibility.
An in silico signal peptide (SP) screening approach based on predicted mRNA structure was designed
to identify new SPs to improve BMP2 secretion. SPs from the TGF-β superfamily were screened, with
the top five candidates and two controls taken forward to in vitro testing in two cell lines. The
approach showed weak predictive power in one line, but none of the tested sequences were able to
induce a significant increase in BMP2 secretion or induce osteogenesis.
These results suggest that substantial further work would be required to create a P11-4-based GAM,
the results presented here indicate that porosity and bioactivity are particularly important. The in silico
SP optimisation approach successfully identified several SPs never before used for SP optimisation,
but required improvement to generate biologically significant predictions.