BACKGROUND: Schwann cells are the most commonly used cells for tissue-engineered nerves. However, autologous Schwann cells are of limited use in a clinical context, and allogeneic Schwann cells induce immunological rejections. Cells that do not induce immunological rejections and that are relatively easy to acquire are urgently needed for transplantation.OBJECTIVE: To bridge sciatic nerve defects using tissue engineered nerves constructed with neural tissue-committed stem cells (NTCSCs) derived from bone marrow; to observe morphology and function of rat nerves following bridging; to determine the effect of autologous nerve transplantation, which serves as the gold standard for evaluating efficacy of tissue-engineered nerves.DESIGN, TIME AND SETTING: This randomized, controlled, animal experiment was performed in the Anatomical laboratory and Biomedical Institute of the Second Military Medical University of Chinese PLA between September 2004 and April 2006.MATERIALS: Five Sprague Dawley rats, aged 1 month and weighing 100-150 g, were used for cell culture. Sixty Sprague Dawiey rats aged 3 months and weighing 220-250 g, were used to establish neurological defect models. Nestin, neuron-specific enolase (NSE), glial fibrillary acidic protein (GFAP), and S-100 antibodies were provided by Santa Cruz Biotechnology, Inc., USA. Acellular nerve grafts were derived from dogs.METHODS: All rats, each with 1-cm gap created in the right sciatic nerve, were randomly assigned to three groups. Each group comprised 20 rats. Autograft nerve transplantation group: the severed 1-cm length nerve segment was reverted, but with the two ends exchanged; the proximal segment was sutured to the distal sciatic nerve stump and the distal segment to the proximal stump. Blank nerve scaffold transplantation group: a 1-cm length acellular nerve graft was used to bridge the sciatic nerve gap. NTCSC engineered nerve transplantation group: a 1-cm length acellular nerve graft, in which NTCSCs were inoculated, was used to bridge the sciatic nerve gap.MAIN OUTCOME MEASURES: Following surgery, sciatic nerve functional index and electrophysiology functions were evaluated for nerve conduction function, including conduction latency, conduction velocity, and action potential peak. Horseradish peroxidase (HRP, 20%) was injected into the gastrocnemius muscle to retrogradely label the L4 and L5 nerve ganglions, as well as neurons in the anterior horn of the spinal cord, in the three groups. Positive expression of nestin, NSE, GFAP, and S-100 were determined using an immunofluorescence double-labeling method.RESULTS: NTCSCs differentiated into neuronal-like cells and glial-like cells within 12 weeks after NTCSC engineered nerve transplantation. HRP retrograde tracing displayed a large amount of HRP-labeted neurons in L4-5 nerve ganglions, as well as the anterior horn of the spinal cord, in both the autograft nerve transplantation and the NTCSC engineered nerve transplantation groups. However, few HRP-labeled neurons were detected in the blank nerve scaffold transplantation group. Nerve bridges in the autograft nerve transplantation and NTCSC engineered nerve transplantation groups exhibited similar morphology to normal nerves. Neither fractures or broken nerve bridges nor neuromas were found after bridging the sciatic nerve gap with NTCSCs-inoculated acellular nerve graft, indicating repair. Conduction latency, action potential, and conduction velocity in the NTCSC engineered nerve transplantation group were identical to the autograft nerve transplantation group (P>0.05), but significantly different from the blank nerve scaffold transplantation group (P