Bone tissue engineering (BTE) holds promise for managing the clinical problem of large bone defects. However, clinical adoption of BTE is limited due to limited vascularization of constructs, which could be circumvented by pre-cultivation of osteogenic and endothelial derived cells in natural-based polymer scaffolds. However, until now not many studies compared the effect of mono- and cocultures pre-seeded in collagen before implantation. We utilized a mouse calvarial defect model and compared five groups of collagen scaffolds: a negative control of a collagen scaffold alone, a positive control treated with BMP-7, monocultures of either human osteoblasts (hOBs) or CD34+ cells, and a coculture of hOB and CD34 cells. Each pre-seeded collagen scaffold was implanted in mice. After 6 weeks mice were sacrificed and their skulls prepared for volumetric and histologic analysis. We found that a monoculture of CD34 cells and a coculture of hOB and CD34 cells pre-cultured in the collagen scaffold increased bone regeneration to a similar extend. In these groups, greater amounts of new bone were found compared with hOB monocultures. Interestingly, monoculture of CD34 cells demonstrated better fracture healing than monoculture of hOBs, emphasizing the possible role of angiogenesis. Our results are promising regarding a cellular based collagen BTE construct, but more work is needed to understand the complex interaction between the osteogenic and endothelial cells. © 2017 Wiley Periodicals, Inc. J Biomed Mater Res Part B: Appl Biomater, 106B: 1505-1516, 2018.