Objective: To investigate the effect of cytoskeleton-associated protein 4 (CKAP4) gene knockout on maxillary expansion osteogenesis and its regulatory mechanism on the osteogenic differentiation of bone marrow mesenchymal stem cells (BMSC). Methods: Sixteen wild type (WT) and sixteen CKAP4 gene knockout (Ckap4-/-) mice aged 6-8 weeks were selected to establish a mouse model of rapid maxillary expansion. Samples were taken on the 7th and 14th day after the operation. Micro-CT and HE staining were used to evaluate bone regeneration. Tissue proteins in the modeled area were collected, and Western blotting analysis (WB) was used to detect the protein expression levels of alkaline phosphatase (ALP), Runt-related transcription factor 2 (RUNX2), and osteocalcin (OCN). BMSC were isolated from WT and Ckap4-/- mice. The expression of surface markers CD29, Sca-1, CD44, CD45, CD34, and CD11b was detected by flow cytometry, and cell proliferation ability was detected by EdU. After 7 days of osteogenic induction, real-time fluorescence quantitative PCR (RT-qPCR) and WB were used to detect the expression levels of RUXN2, ALP, OCN, protein kinase B (AKT), and phosphorylated protein kinase B (p-AKT). After 21 days, alizarin red staining and cetyl pyridine chloride quantification were used to detect the differences in mineralized nodule formation in each group. In CKAP4 gene knockout BMSC, the small-molecule AKT agonist sc79 (4 μg/ml) was added as the intervention group (Ckap4-/- +sc79), and dimethyl sulfoxide (DMSO) treatment was used as the control group (Ckap4-/- +DMSO). After osteogenic induction, RT-qPCR, WB, and alizarin red staining were used to compare the osteogenic differentiation differences between the two groups of cells. Results: The micro-CT results showed that at 7 days and 14 days after surgery, the new bone volume in the Ckap4-/- group [(0.070±0.010) and (0.146±0.019) mm3] was significantly lower than that in the WT group [(0.094±0.006) and (0.196±0.013) mm3] (both P<0.01). HE-stained histological sections showed that the area of new bone tissue in the Ckap4-/- group at 7 days and 14 days after surgery [(0.101±0.008) and (0.158±0.010) mm2] was also significantly lower than that in the WT group [(0.116±0.005) and (0.183±0.008) mm2] (both P<0.05). WB was used to detect the tissue proteins in the maxillary modeling area of mice in the two groups 7 days after surgery. The results showed that the expression levels of ALP, RUNX2 and OCN in the Ckap4-/- group were significantly lower than those in the WT group. BMSC from wild-type mice and CKAP4 knockout mice were both positively expressed for CD29, CD44, and Sca-1, and basically not expressed for CD45, CD34, and CD11b. EdU assay showed that there was no significant difference in the proliferation ability of cells in the two groups. After 21 days of osteogenic induction of BMSC, alizarin red staining results showed that the number of mineralized nodules in the Ckap4-/- group was significantly less than that in the WT group. After adding sc79, the number of mineralized nodules increased significantly, which was consistent with the results of cetyl pyridine chloride quantification. After 7 days of osteogenic induction, of ALP, RUNX2, and OCN It was found that the expression levels in the CKAP4-/-group (0.751±0.066, 0.484±0.040, 0.679±0.063) were significantly lower than those in the WT group (1.000±0.113, 1.000±0.081, 1.000±0.113) (all P<0.001). The results of WB were consistent with those of RT-qPCR. At the same time, the WB results showed that the level of p-AKT protein in the CKAP4-/-group (0.518±0.114) was significantly lower than that in the WT group (1.000±0.234) (P<0.05). After treatment with sc79 for 7 days of osteogenic induction, RT-qPCR was used to detect the gene expression levels of ALP, RUNX2, and OCN. The results showed that the expression levels in the CKAP4-/-+sc79 group (2.755±0.353, 4.800±0.990, 2.524±0.137) were significantly higher than those in the CKAP4-/-+DMSO group (1.000±0.078, 1.000±0.247, 1.000±0.175) (all P<0.001). Conclusions: CKAP4 knockout inhibits the osteogenic differentiation of BMSC by reducing the activity of the PI3K/AKT signaling pathway, thereby suppressing osteogenesis in maxillary expansion.