Expression, Purification and Charaterization of Recombinant Human Bone Morphogenetic Protein 2

Abstract

Bone Morphogenetic Proteins (BMPs) are categorized as subfamily of Transforming Growth Factor beta and perform important functions in diverse cellular processes from embryo to adult life. They are important regulators of bone morphogenesis and proved their therapeutic potential in spine fusion and ortho/maxillofacial surgeries. Two members of the BMP family, that is BMP2 and BMP7 had been approved by FDA as recombinant therapeutic drugs for human use. Normally, BMPs are active in low doses (5-20ng/ml) but commercially available are used in high doses, up to 40mg in an in-vivo translational procedure. The high dose is required due to the short systemic half-life of BMP2 and single dose is not adequate to stimulate the osteo-inductive response. Currently, the commercially available recombinant human BMP2 (rhBMP2) is been produced from mammalian expression system mainly CHO and HEK cell lines, with a collagen sponge as a carrier and is highly expensive. Despite been having advantage of biologically active protein, mammalian expression system is associated with high cost of production and low yield. Alternate approaches for the cost effective recombinant production of proteins for the successful therapeutic applications is the prokaryotic expression system. In prokaryotic system, B. subtilis and E. coli are the two most widely used expression hosts. B. subtilis has the advantage of secreting proteins directly in the culture medium and the production of BMP2 had not been reported previously. Whereas, E. coli has been known for the high yield but major disadvantage of E. coli is that, majority of the heterologous proteins are produced in an inactive form as inclusion bodies and BMPs had always been produced recombinantly in this way which were refolded in-vitro. So, in case of E. coli, soluble production with SUMO fusion technology is also a novelty in the recombinant production of native BMP2. In the current study, for the cost effective production of rhBMP2 in B. subtilis expression system, BMP2 gene was cloned in the form of monomer and a covalently linked homodimer containing glycine-serine rich linker into pHT43, B. subtilis expression vector. After confirmation of positive clones with colony PCR and restriction analysis, sequencing was performed to rule out any possible mutation during amplification. The recombinant plasmids were transformed in SCK6 and WB600 strains of the B. subtilis for the secretory expression. The expression was optimized in three different media that is LB, 2xYT and 2X LMAL at 30 and 37°C. Maximum expression of ~35% of total secretory protein for rhBMP2 monomer was observed at 30°C in 2xYT medium with SCK6 strain. The optimized conditions for IPTG concentration and fermentation time was 0.6mM of IPTG for 20 hours of fermentation. Auto induction with lactose was also optimized for large scale production in bio-fermenter and maximum expression was observed at 6mM of lactose for 24 hours of fermentation. The rhBMP2 homodimer was grown at pre-optimized conditions of rhBMP2 monomer for media, strain and temperature. However, IPTG concentration and fermentation time was optimized separately. Maximum expression of ~25% of total secretory protein was observed at 0.8mM of IPTG for 8 hours of fermentation and with lactose the optimized conditions were 10mM of lactose for 16 hours of fermentation. The rhBMP2 monomer and homodimer were characterized by western blotting with anti-human BMP2 antibody and Native PAGE analysis for confirmation of the homodimer. Both the protein were then purified with anion exchange chromatography using FPLC system. The specific elution concentration for rhBMP2 monomer and homodimer were 0.9M and 0.6M of NaCl respectively. The final yield of 1mg and 1.84mg with 90 and 80% purity for rhBMP2 monomer and homodimer respectively was achieved from 200ml of culture supernatant. The purified proteins were then further characterized for their biological activity by alkaline phosphatase (ALP) assay on C2C12 mouse myoblastic cell line. Different doses of rhBMP2 monomer and homodimer i.e. 0, 50, 100, 200 and 400 ng/ml were tested and maximum ALP activity was observed at 200ng/ml. Results showed dose dependent increase in the ALP activity and a decline in the activity was observed at higher doses. The rhBMP2 standard was used as positive control whereas, DMEM medium without rhBMP2 was used as negative control. Our results showed that biological active rhBMP2 has been produced as a secretory protein in B. subtilis and in dimeric configuration. However, the purification of the protein from large volume of culture supernatant was difficult and the final yield was low. So, the system was not considered to be cost effective for bulk production. As an alternate strategy, E. coli expression system was used for cost effective production of rhBMP2. Previously rhBMP2, was produced in inclusion bodies as biologically inactive protein which was refolded in-vitro. So, for the soluble expression of rhBMP2, SUMO fusion technology was opted and a fusion gene was constructed with human SUMO3 fused to the N-terminal of human BMP2. The fused gene was cloned in pET21a (+) expression vector and confirmed with restriction analysis and sequencing. The recombinant vector was transformed into Rosetta gami B (DE3) and BL21 codon plus strains of E. coli for expression analysis. Results showed the soluble expression of fusion protein in the cytoplasmic soluble fraction after sub cellular fractionation of total cell protein. Optimizations were performed for maximum soluble expression and ~40% expression was achieved at 37°C in LB medium with BL21 codon plus strain. The IPTG and lactose concentrations were also optimized and maximum expression was observed with 0.3mM of IPTG at 6 hours of fermentation and 6mM of lactose for 8 hours of fermentation. The fusion protein was characterized by western blotting and native PAGE showing positive confirmation of expression with anti-human BMP2 antibody and dimeric nature was confirmed by native PAGE. Cleavage of SUMO tag, resulted in the aggregation of the rhBMP2 and converted it into an insoluble form again. But, the fusion protein was proceeded for purification with anion exchange chromatography and eluted at 0.5M concentration of NaCl. Final yield of ~75mg/L with 75% purity and 30% recovery was achieved. Further, the fusion protein was tested for biological activity by induction in C2C12 cell and showed dose dependent increase in the ALP activity with maximum at 200ng/ml. The computational analysis was performed and the 3D structure of the fusion protein was constructed with MODELLER and validated with Molecular Dynamic (MD) simulation for 20 ns. Results showed that the constructed model was stable and reliable as shown by the negligible fluctuations in RMSD values in the last 10 ns of MD simulation. The predicted model was further tested for interaction with BMP type-I receptor by using HADDOCK webserver. The best model with lowest HADDOCK score was superimposed with actual crystal structure of BMP2 and BMPR1A and showed minor RMSD variations. Furthermore, the ligand and interface RMSD values of the HADDOCK generated model were in the acceptable range of ~2-5Å with the cut off value of 10Å. The computational analysis justified the biological activity of the rhBMP2 in fusion with human SUMO3 showing no hindrance of SUMO3 in the binding of rhBMP2 with its receptor. In conclusion, B. subtilis had the advantage of fully functional bioactive production of rhBMP2, but further optimizations in the purification strategy is required to reduce the cost of production. Furthermore, optimizations in the selection of promoter and signal peptide could increase the yield. Whereas, in E. coli system, the final yield was high as compared to the Bacillus system but it requires fusion partner for soluble production of rhBMP2. In the current study, removal of SUMO3 caused rhBMP2 to form aggregates. Further, investigations are required in the mechanism of SUMO mediated solubilization with advanced techniques to rule out if SUMO is making modifications in the rhBMP2 structure or acting just as a carrier. Moreover, other fusion tags could be tested for the soluble production. For future directions, BMP2 variants with enhanced binding affinities can be generated to reduce the effective dose of concentration