Introduction Bone marrow mesenchymal stem cells (MSCs) have been used in regenerative bone biology for more than a decade, as they can be easily recovered from patients. More recently some of the focus in this field has shifted towards the use of embryonic stem cells (ESCs). Previous studies reported that ESCs can be induced to differentiate into cells showing a mature osteoblastic phenotype by culturing them under osteoinductive conditions. This study proposes determination of the bone regeneration potential of murine bone marrow MSCs and ESCs. Materials and methods For this purpose, a murine model (dual transgenic mice αSMARFP/Col2.3GFP) has been utilized. After isolation of cells from the bone marrow of long bones, MSCs have been identified through the expression of alpha-smooth muscle actin (αSMA) promoter directed RFP. To track the transition of MSCs into mature osteoblasts, a bone-specific fragment of rat type I collagen promoter driving green fluorescent protein (Col2.3GFP) has been utilized. In addition, ESC lines have been derived from the same αSMARFP/Col2.3GFP transgenic mice, allowing identification of cells at the mesenchymal stage (αSMARFP) and at mature osteoblast stage (Col2.3GFP). In vitro analysis of osteogenic potential has been made for both types of cells and in vivo analysis (teratoma assay) of osteogenic potential for ESCs. Results Cultures of bone marrow MSCs were established, and after one week a population of αSMARFP expressing cells was noticed. Following cell sorting, replated αSMARFP+ cells (20-30%) were induced to osteogenesis, and strong expression of osteoblast specific transgene Col2.3GFP was noticed, confirming osteogenic ability of αSMARFP+ cells. Osteogenic differentiation was confirmed by detection of mineralization with von Kossa method and by up-regulation of markers of mature osteoblast lineage cells; osteocalcin and bone sialoprotein. Following osteogenic induction in ESCs, expression of alkaline phosphatase and subsequent mineralization (detected by von Kossa staining) was observed. After one week of osteogenic induction, ESCs began to express αSMARFP. This expression was localized to the peripheral area encircling a typical ESC colony. Nevertheless, these αSMARFP+ cells did not show activation of the Col2.3GFP, even after 6 weeks of osteogenic differentiation in vitro. Also, small increases in expression in some of the bone markers analyzed (Colagen type I, bone sialoprotein) were noticed, but most of them appeared minimal compared to the levels expressed by the MSCs, even after six weeks of osteogenic induction. In contrast, in vivo teratoma assay by ECSs showed bone formation and strong Col2.3GFP signal in the areas that stain positive for mineralization by von Kossa. Conclusion These results show that MSCs compare to ESCs, have much more capability to differentiate into mature osteoblast, express specific bone markers and Col2.3GFP transgene after osteogenic induction in vitro. On the other hand, in the functional in vivo assay ESCs show capability to form bone and activate Col2.3GFP marker. The results obtained indicate that detection of alkaline phosphatase activity and mineralization of ESCs cultured under osteogenic conditions is not sufficient to demonstrate osteogenic maturation. This study indicates the utility of the promoter-visual transgene approach to assess the commitment and differentiation of stem cells into the osteoblast lineage. To current knowledge, this study is the first to utilize the expression of Col2.3GFP transgene in the ESCs derived from transgenic mice.