Induction of bone formation in abdominal implants constituted by collagen sponges embedded with plant-based human transforming growth factor family proteins in ectopic dog model

Abstract

Background: Trauma, osteomyelitis, bone tumour resections and congenital deformities are the main causes of bone deficiency in which autologous graft is the preferred treatment, but usually the bone supplies are limited.

Methods: An experimental model of heterotopic bone formation in the subcutaneous abdominal area of dogs was developed. This model consists in omentum wrapped implants constituted by collagen type 1 sponges embedded with demineralized bone powder, calcium cloride, thrombin and platelet rich plasma; the implant is totally converted in trabecular bone after four months of implantation. This model was improved by accelerating bone production, after the isolation of the most conspicuous histological constituents (inflammatory, bone and adipose tissues) by laser microdisection and purified from them RNA that was used to determine by RT-PCR the gene expression kinetics of the most important growth bone factors. Then, the most abundant and rapidly synthesized factors were produced by genetic engineering in tobacco plants.

Results: Bone morphogenetic proteins 2 and 7 and transforming growth factor-β1were the most rapidly and highly synthesized factors, and they were efficiently produced in a genetic engineering plant based system in tobacco leaves. Their incorporation as recombinant proteins in the scaffold collagen sponge induced in just one month mature heterotopic bone.

Conclusion: This study demonstrates for the first time that this plant system is able to produce recombinant bone growth factors in high amount and at low cost, and they were highly efficient to rapidly induce bone formation in abdominal implants potentially useful for autotransplantation.

Keywords: Bone morphogenetic proteins; Heterotopic bone; Tissue engineering.

Figure 1

Kinetics of bone growth factors gene expression during heterotopic bone formation. Bone trabeculae (OST), inflammatory tissue (INF) and adipose tissue (ADP) were isolated from the implants using laser microdisection and used to isolate total RNA to determine by RT-PCR the indicated bone growth factor during the heterotopic bone formation. Numbers indicate the day when the implant was obtained after the implantation. Results are present as mean and standard deviation from three dogs in the indicated time points. Two independent experiments were run with similar results. Asterisks represent statistical significance comparing the groups in each time point (p < 0.05).

Figure 2

Identification by western-blot of recombinant BMP-2, BMP-7 and TGF-β1 from Nicotiana benthamiana leaves. Crude extracts from plants containing each factor and obtained as described in the text were filtered on 0.2 μm membranes and loaded on a HisTrap HP column. Each factor was loaded separately. Fractions were dialyzed against PBS, loaded on a SDS-PAGE gel, transferred to nitrocellulose membranes and incubated with specific rabbit policlonal antibodies to confirm its identity. Molecular weight markers (kDa) are indicated on the left.

Figure 3

Determination of the percentage of bone tissue by automated morphometry after one and four weeks of implantation adding the indicated growth factors. Results are presented as means and standard deviation of three different dogs in the indicated time points. Mix means implants with the three factors in the same concentration. Asterisk represent statistical difference (p < 0.005) among the indicated groups. Two independent experiments were run with similar results.

Figure 4

Representative histology of implants after 4 weeks of abdominal implantation. A) Control implant without any factor is essentially constituted of fibroadipose tissue with small areas of bone (arrows). B) Fibrous tissue in control implant is well demonstrated by Masson trichromic staining. C) In contrast, implant with recombinant BMP-2 shows large mature bone trabeculae surrounded by fibrous or mesenchymal tissue. D) This mesenchymal tissue shows numerous cells with PCNA positive immunostaining. E) Similar bone production is observed in implant with BMP-7. F) This implant with BMP7 shows extensive areas of osteoid in section stained with Masson technique (asterisk). G) Even more bone trabeculae are seen in an implant with addition of TGF-1. H) They are mature trabeculae constituted by well calcified bone as show by Von Kossa staining. I) In comparison, fewer bone trabeculae were induced in implant with all the TGF family factors.