OBJECTIVE Construction of viable tissue engineered bone is one of the most important research fields in the clinical application of bone tissue engineering, to investigate the function of nerve factors in bone tissue engineering by cell detection in vitro and construction of neurotization tissue engineered bone in vivo. METHODS Fifty-four healthy New Zealand white rabbits, male or female, weighing 2-3 kg, were involved in this study. Bone marrow mesenchymal stem cells (BMSCs) from the bone marrow of white rabbits were cultured. The second passage of BMSCs were treated with sensory nerve or motor nerve homogenates, using the LG-DMEM complete medium as control. The proliferation and osteogenic differentiation of the cells were observed and tested by the MTT assay, alkaline phosphatase (ALP) stain, and collagen type I immunocytochemistry identification. The osteogenic induced BMSCs were inoculated in beta tricalcium phosphate (beta-TCP) biomaterial scaffold and cultured for 72 hours, then the beta-TCP loaded with seed cells was implanted in the rabbit femur with 15 mm bone and periosteum defects. Fifty-four New Zealand white rabbits were randomly divided into three groups (n = 18): sensory nerve bundle (group A) or motor nerve bundle (group B) were transplanted into the side groove of beta-TCP scaffold, group C was used as a control without nerve bundle transplantation. X-ray detection was performed at the 4th, 8th, and 12th weeks after operation. Bone mineral density (BMD) detection and S-100, calcitonin gene-related peptide (CGRP) immunohistochemistry stain were used at the 12th week to evaluate the effects of bone formation and discuss the mechanism. RESULTS MTT assay indicated that the absorbance (A) value of each group increased with culture time. From the 6th day, the A values of both the sensory nerve and motor nerve homogenate groups were lower than that of the control group, showing significant difference (P < 0.01). On the 8th and 10th days, the A value of the sensory nerve homogenate group was lower than that of the motor nerve homogenate group, showing significant difference (P < 0.05). ALP stain and collagen type I immunocytochemistry identification indicated that the positive cells in both the sensory nerve and motor nerve homogenate groups were less than that of control group after culturing 7 days. And the positive expression of collagen type I was just visible in the cells of control group. The Yang's scores increased gradually in three groups, the score of group A was significantly higher than those of group B and group C (P < 0.01) at the 8th week. The BMD value of group A was significantly higher than those of group B and group C (P < 0.01) at the 12th week. The S-100 and CGRP expressions were high in group A, and low in group B and group C. CONCLUSION Homogenates of sensory nerve and motor nerve have inhibitory effects on the proliferation and osteogenic differentiation of BMSCs. The osteogenesis and remodeling of the neurotization tissue engineered bone are more closely related with sensory nerves.