This study investigated whether gene transfer to the tendon-bone insertion site is possible during early tendon-transplant healing using viral vectors. In addition, we evaluated the optimal gene delivery technique for an in vivo adenoviral gene transfer to a tendon-bone insertion site in a bone tunnel. Twenty-six rabbits underwent a bilateral transfer of the flexor digitorum longus tendon into a bone canal in the calcaneus. The animals were divided into two groups. The first group (n=18) received a direct injection of an adenoviral vector carrying the luciferase marker gene into the tendon on the left side, while on the right side the adenoviral vector was first injected into the bone trough and the tendon was later inserted into the trough. The analysis of this experiment showed that over a 4-week period a higher luciferase activity was achieved using the bone trough immersion technique. In the second group (n=8) we therefore used the qualitative marker virus (Ad/-LacZ) with the bone trough immersion technique in order to show the site of gene expression. The histological analysis of this experiment demonstrated the presence of beta-galactosidase positive cells within the tendon-bone interface over a 4-week period. Therefore we showed in the first part of this study that the bone canal provides a more efficient target for direct adenoviral gene delivery than the tendon. In the second part of the study we demonstrated the feasibility of the bone trough immersion technique since sustained gene expression within the tendon-bone interface was obtained for up to 4 weeks. This study has shown the feasibility of gene delivery to the tendon-bone interface and provides the basis for the application adenoviral delivery of growth factor genes to the tendon-bone insertion site.