gms | German Medical Science

Objectives: Two critical events early in fracture healing are thought to determine the course of repair: 1) lack of oxygen caused by blood vessel rupture and 2) mechanical instability. Our goal is to define how mechanical loading alters the identity and distribution of hypoxic cells in early fracture repair. Toward this end, we here validated a methodological pipeline to precisely delineate cellular oxygen concentrations in the bone marrow and fracture gap. Nitroimidazoles such as EF5 are selectively reduced by nitroreductase enzymes under hypoxic conditions, resulting in the formation of EF5 adducts that can be visualized with fluorophore-coupled antibody. Here, we establish EF5 staining as a precise method to delineate cells by levels of intracellular oxygen.

Methods: First, we isolated bone marrow cells from femora of C57BL/6J female mice and performed cultivation under defined oxygen concentrations (10%, 2%, 0.5%, 0.1% pO2) for 2h with supplementation of EF5 in the cell culture medium. Cells were fixed with 4% paraformaldehyde directly after collection, stained overnight with Cy5-conjugated anti-EF5 antibody and analyzed using flow cytometry. Next, 12–14 week-old C57BL/6J female mice underwent femoral osteotomy (fixation with external fixator) and were injected with 10 mM EF5 and euthanized after 2 hours at 3 and 5 days post-surgery. Bone marrow from naïve mice was used for verification and control. Femora were either directly cryo-embedded or flushed to isolate the respective cells. EF5 staining was performed as verified before distinguishing cells from the fracture gap, ipsilateral and contralateral bone marrow. All procedures were conducted in accordance with UPenn IACUC regulations (protocol no: 806482).

Results and conclusion: To validate the EF5 staining method, murine bone marrow cells were cultivated under different oxygen concentrations resulting in the presence of a EF5+ and EF5- cell fraction at 10% pO2 and almost exclusively EF5+ cells at 0.1% oxygen. Frequency quantification of naïve bone marrow cells revealed that 59.4±13% were positive for EF5 while 40.64±13% of cells were not positive for EF5. Immunofluorescence staining of the whole bone marrow identified regions with high and low EF5 signal intensity, indicating heterogenous intracellular hypoxia. Finally, we did not find any EF5+ cells in the fracture gap when compared to the ipsilateral and contralateral bone marrow indicating higher local oxygen concentrations in the fracture gap at 3d and 5d post-fracture. In summary, our results indicate a substantial heterogeneity in intracellular hypoxia (equivalent to EF5 staining intensity) among bone marrow cells in situ but also when cultivated in vitro under defined oxygen concentrations. Moreover, we provide evidence that the initial phase of fracture healings is not hypoxic. Future studies are needed to delineate the underlying mechanism and immediate consequences for the regenerative process.