Background: Bone morphogenetic proteins (BMPs) are now being used as bone-graft substitutes to enhance spinal fusion. However, the large doses of BMP required to induce a spinal fusion in humans suggests that the delivery of these proteins should be improved. We used ex vivo adenoviral gene transfer to create BMP-2-producing bone marrow cells, and these autologous cells were found to induce a posterolateral fusion of the spine in syngeneic rats.
Methods: Intertransverse spinal arthrodesis (L4 and L5) was attempted in ten groups of Lewis rats with 5 × 10
6 BMP-2-producing rat bone marrow cells (Ad-BMP-2 cells), created through adenoviral gene transfer with guanidine hydrochloride-extracted demineralized bone matrix as a carrier (Group I); 5 × 10
6 Ad-BMP-2 cells on a collagen sponge carrier (Group II); 10 µg of recombinant BMP-2 (rhBMP-2) in a guanidine hydrochloride-extracted demineralized bone matrix carrier (Group III); 10 µg of rhBMP-2 in a collagen sponge carrier (Group IV); autogenous iliac crest bone-grafting (Group V); 5 × 10
6 ß-galactosidase-producing rat bone marrow cells, created through adenoviral gene transfer with guanidine hydrochloride-extracted demineralized bone matrix as a carrier (Group VI); decortication of the transverse processes alone (Group VII); 5 × 10
6 uninfected rat bone marrow cells with a guanidine hydrochloride-extracted demineralized bone matrix carrier (Group VIII); guanidine hydrochloride-extracted demineralized bone matrix only (Group IX); or a collagen sponge alone (Group X). Each specimen underwent plain radiography, manual palpation, and histological analysis.
Results: All spines in Groups I and II (BMP-2-producing bone marrow cells) and all spines in Groups III and IV were fused at four weeks postoperatively. In contrast, none of the spines in the other groups had fused at a minimum of eight weeks after implantation. Histological analysis of the specimens revealed that the spines that had received BMP-2-producing bone marrow cells (Groups I and II) were filled with coarse trabecular bone postoperatively, whereas those that had received rhBMP-2 (Groups III and IV) were filled with thin, lace-like trabecular bone. All of the other spines, including those that had been treated with autogenous iliac crest bone-grafting (Group V), produced little or no new bone.
Conclusion: BMP-2-producing bone marrow cells, created by adenoviral gene transfer, produce sufficient BMP to induce an intertransverse fusion in the rat spine model.
Clinical Relevance: Regional gene therapy can be used to induce spinal fusion. This strategy with use of transduced bone marrow cells created through ex vivo gene transfer with a BMP-2-containing adenovirus could be adapted to enhance spinal fusion in humans.