Osteocytes make up >90% of bone cells, are embedded in mineralised matrix where they form a communication network. Osteocytes differentiate from osteoblasts, and are thought to be mechano-sensitive. They are very difficult to isolate leading to a dependence on cell lines for in vitro studies of osteocyte biology. There is thus a need to develop new methods to study these cells. Recent publications indicate that osteoblasts maintained in in vitro 3D collagen gels may differentiate to osteocytes.
We maintained osteoblasts (MC-3T3; human primary) in 3D type I collagen gels (250 μl; 48-well plates) for 15 days in either α-MEM (basal medium), or mineralising medium (basal medium, dexamethasone, β-glycerophosphate). Cell number, viability and phenotype (IHC, qRTPCR, confocal microscopy), gel stiffness (Losenhausen), and VEGF and IL6 secretion (ELISA) were quantified.
Cells appeared more dendritic over time and formed connecting cellular networks (H&E, phalloidin). Cell viability was similar in both media (>85% MC-3T3s; >95% human primary), but cell numbers were significantly higher (P<0.001) in mineralising conditions. Calcein and alizarin red staining confirmed the presence of mineralisation from day 7. DMP-1 (osteoblast/osteocyte differentiation marker) was not expressed (IHC) at day 3 but then gradually increased in expression (days 714). E11 (osteocyte marker localised to dendrites) was low at day 3 (IHC, qRTPCR), peaked at day 10, and returned to lower levels by day 14. Gel stiffness significantly increased over 11 days (P<0.01) and the mineralised gels were stiffer than those in basal medium (P<0.01). VEGF and IL6 secretion also changed significantly with time and culture conditions.
Osteoblasts maintained in 3D gels appear to differentiate along the osteocytic pathway. It is possible to mineralise these cultures thus mimicking further their in vivo environment. This methodology provides a novel model to study osteocyte biology, and will enable studies relating to bone loading, repair and regeneration.
Declaration of funding: Yes.
Details: This work was supported by Arthritis Research UK.