Team:MIT mammalian Bone


Bone Formation


The goal of this portion of the project was to synthetically differentiate stem cells into bone; this would act as the 'output' for our final system. Since stem cells can by tricky to genetically engineer, we chose to build our circuit in human endothelial kidney (HEK) cells; we also considered using chinese hamster ovary (CHO) cells, which adhere better to the seeding substrate and might be more robust to mechanical stimulation. In the final system, HEK cells will secrete a diffusible morphogen to differentiate co-cultured stem cells. We used Bone Morphogenetic Protein 2 (BMP2) as our osteogenic signaling molecule; it is one of the central regulators of osteoblast differentiation in mammalian cells (1). Upon binding BMP2, the BMPRI receptor activates a downstream signaling cascade involving Smad proteins, which results in activation of the Runt-related transcription factor 2 (RUNX2), the main transcription factor controlling osteoblast differentiation. BMP2 has been shown to induce transdifferentiation in multiple stem cell types (2, 3, 4). Here, we work with myoblastic progenitor (C2C12) and mesenchymal 2 (C3HT101/) stem cells, two cell lines which have been shown capable of osteoblast differentiation (5, 6).

In Vitro Osteoblast Differentiation

Our first task was to see if we could induce osteogenic differentiation in our stem cell lines. We obtained a stock of human recombinant BMP2 protein and added it to the supernatant of stem cell cultures at 100ng/ml and 300ng/ml, based on a review of concentrations used in the literature. We tested the response to both constitutive (all 5 days) and transient (induction stopped on day 2) BMP2 signaling. HEK cells were also tested as a negative control for differentiation. Micrographs were taken on day 5 to assay for changes morphology. Click here for a detailed experimental writeup.

Morphology Results

Figure 1. Brightfield Microscopy Taken Prior to ALP Assay. Both C2C12 and C3HT101/2 cell lines showed significant morphological change in response to BMP2.

Morphology Results Summary

(Figure 1) Both C2C12 and C3HT101/2 were affected by the addition of BMP2. C3HT101/2 exhibited the strongest response; at 300ng/ml of BMP2, we observed small mineralized nodules beginning to form, as previously described in the literature. However there was negligible change in morphology of the transiently stimulated C3HT101/2 cells, indicating that BMP2 induction is required over a longer time period to induce differentiation. The C2C12 uninduced cells began differentiation into muscle tissue, seen in multinucelated myofibril formation; the C2C12 cells induced with BMP2 instead began to exhibit morphological osteoblast markers. We observed no significant changes in HEK cell morphology

ALP Assay

The next step was to quantify our results. We preformed an Alkaline Phosphatase (ALP) assay on the cultures, a routine protocol used to determine differentiation. It measures activity of an osteoblast-specific protein, alkaline phosphatase. This enzyme increases the local concentration of phosphate, which aids formation of the the hydroxyapatite ion that underlies mineralization in bone.

Figure 2. Photograph of ALP Assay Result.

ALP Assay Results

(Figure 2) C2C12 displayed a strong increase in alkaline phosphatase activity in cells treated with 100ng/ml and 300ng/ml BMP2, when compared with untreated controls. The C3HT101/2 mesenchymal stem cells exhibited a significantly weaker, but still visible upregulation of ALP in cells treated with BMP2. HEK and transiently stimulated C3HT101/2 did not display visible upregulation of ALP activity.

Co-Culture Assays

To test the viability of our final system, we grew co-cultures of our stem cell lines with HEK and CHO cells. Below are micrographs of the co-cultures of at the indicated ratios; we cultured HEK with MSC (C3HT101/2 mesenchymal stem cells) and C2C12, and CHO with C2C12. Cells were seeded at a total density of 200,000 cells/well in 6-well plates. All images were collected on Day 6.

Figure 3. Brightfield Microscopy of Co-Cultures.

Co-Culture Results (Figure 3) Cells grew to confluency; we did not observe any incompatibility issues between different cell types. Interestingly enough, MSC cells grew naturally together to form 'clumps' in the midst of a sea of HEK cells. We observed myoblast differentiation of C2C12 cells; at higher seeding ratios in HEK cells, the C2C12 cells began to form the myotubules. This effect was not observed when co-cultured with CHO cells.


We have demonstrated osteoblast differentiation in two different stem cell lines, using human recombinant BMP2. We have also shown here a proof of concept for our final system, a co-culture of two different cell types. Although not shown here, we have also constructed cell lines containing BMP2 under the control of an inducible and constitutive promoter. We plan to begin functionally testing these cell lines in the coming weeks. References

1. Hogan, B. L. (1996) Harvey Lect. 92, 83-98

2. Katagiri, T., Yamaguchi, A., Komaki, M., Ab, E., Takahashi, N., Ikeda, T., Rosen, V., Wozney, J. M., Fujisawa-Sehara, A., Suda, T. (1994) J. Cell Biol. 127, 1755-1766

3. Katagiri T, Yamaguchi A, Ikeda T, Yoshiki S, Wozney JM, Rosen V, Wang EA, Tanaka H, Omura S, Suda T (1990) . Biochem Biophy Res Commun 172:295–299

4. Yamaguchi A, Katagiri T, Ikeda T, Wozney JM, Rosen V, Wang EA, Kahn AJ, Suda T, Yoshiki S (1991) . J Cell Biol 113:681–

5. Nishimura, R, Kato Y, Chen D, Harris SE, Mundy GR, and Yoneda T (1998) J Biol Chem 273: 1872-1879

6. Richard S, Torabi N, Franco GV, Tremblay GA, Chen T, et al. (2005). PLoS Genet 1(6): e74. doi:10.1371/journal.pgen.0010074