INTRODUCTION: The availability of autogenous tendons (middle part of patellar tendon, semitendinosus/gracilis, or quadriceps tendon) for cruciate ligament reconstructions is restricted and related to withdrawal morbidity. Allografts and synthetic ligament materials often show problems regarding long-term stability and immunological reactions. Therefore, the aim of this study was to develop and characterize a new scaffold based on acellular allografts seeded with autologous cells for tissue engineering of the anterior cruciate ligament (ACL). MATERIALS AND METHODS: Semitendinosus tendons of New Zealand White (NZW) rabbits were harvested and acellularized using the detergent sodium dodecyle sulfate (SDS) as the main ingredient. After that, cultured (37 degrees C, 5% CO(2), medium) dermal fibroblasts were injected into the tendons. These constructs were further cultivated for 4, 7, or 14 days under the same culture conditions. Native, acellular, and seeded tendons underwent biomechanical testing (ultimate load to failure [N], stiffness [N/mm], and elongation [%], each n = 9] and histological hematoxylin-eosin (H.E.) staining. Detailed immunohistochemical (collagen I, III, IV, VI, pro-collagen I, versican, and vimentin) analyses were conducted to detect changes in the composition and structure of the extracellular matrix (ECM) after acellularization. RESULTS: Histologically, a cell-free, crimped slack tendon structure after acellularization and a good integration of the cells after injection (4, 7, and 14 days) were seen. Metabolic activity of the seeded cells was demonstrated by positive immunohistochemical staining for pro-collagen I, which was negative in nonseeded constructs. Major differences in staining patterns of the various other ECM components were not observed. Biomechanically, the maximum load to failure of these tendons was comparable to native tendons (P = 0.429; native 134.5 +/- 12.9 N; acellular 118.5 +/- 7.3 N; seeded 132.3 +/- 5.6 N). Stiffness and elongation were comparable between native and acellular tendons, but differed significantly after seeding (P < 0.001). CONCLUSION: The described method is suitable to make tendons completely cell free without changing their major biomechanical properties. Preservation of the ECM and of the collagen fiber structure by this method should give an ideal environment for autologous cell integration and metabolic activity in contrast to other approaches for tissue acellularization. The cell disruption and extraction of cell detritus should minimize adverse immunogenic reactions.