Study Design. We demonstrated the differentiation of notochordal cells by direct observation using a live automated cell imaging system. We also hypothesized that notochordal cells have characteristics of chondrocyte-like cells. Objective. To determine characteristics of notochordal cells by matrix protein expression and their differentiation using a live automated cell imager. Summary of Background Data. Although notochordal cells are critical to homeostasis of intervertebral disc, their fate has not been extensively studied and there is little evidence of notochordal cells as progenitors. Methods. Notochordal cells purified from rabbit nucleus pulposus were isolated after serial filtration. Notochordal cells in 3-dimensional culture were compared to chondrocyte-like cells by ³⁵S sulfate incorporation into proteoglycan and reverse transcription polymerase chain reaction for gene expression(collagen II and aggrecan). Notochordal cells in 2-D culture were used for immunocytochemical staining (collagen II, aggrecan, and SOX9) and time-lapsed cell tracking study. Results. Notochordal cells were capable of proteoglycan production at a rate comparable to chondrocyte-like cells (108% ± 22.6% to chondrocyte-like cells) and expressed collagen II, aggrecan, and SOX9. In time-lapsed cell tracking analysis, notochordal cells were slower in population doubling time than chondrocyte-like cells and differentiated into 3 morphologically distinct cell types: vacuolated cells (area: 2392 ± 507.1 μm², velocity: 0.09 ± 0.01 μm/min); giant cells (area: 12678 ± 1637.0 μm², velocity: 0.08 ± 0.01 μm/min) which grew rapidly without cell division; polygonal cells (area: 3053 ± 751.2 μm², 0.14 ± 0.01 μm/min) morphologically similar to typical differentiation type of chondrocyte-like cells (area: 2671 ± 235.6 μm², 0.19 ± 0.01 μm/min). Rarely, notochordal cells formed clusters analogous to that observed in vivo. Conclusion. These studies demonstrate a chondrocyte phenotype of notochordal cells and are the first direct evidence of notochordal cell differentiation, suggesting that they may act as progenitor cells, which has the potential to lead to their use in novel approaches to regeneration of degenerative intervertebral disc.

As clinical evidence suggests that mechanical forces can have both reparative and traumatic effects on the spine, we investigated responses to different magnitudes, frequencies, and durations of applied tensile strain in an in vitro system. We examined the interactions of inflammatory and mechanical stimuli on cells isolated from the annulus fibrosus. Rabbit annulus fibrosus fibrochondrocytes were cultured in the presence or absence of an inflammatory stimulus. Cells were exposed to various magnitudes and frequencies of tensile strain for 4 or 24 h, and mRNA expression of catabolic mediators of inflammation and matrix degradation was measured by quantitative real time PCR and compared to control cells. Conditioned media were analyzed for matrix metalloprotease activity and production of prostaglandin E₂. Application of low magnitudes and frequencies of tensile strain resulted in down‐regulation of catabolic mediators, particularly under inflammatory stress. However, loss of this protective effect was observed at higher frequency and magnitude, and after prolonged duration. These in vitro data confirm the existence of magnitude, frequency, and duration based effects, which determine biochemical response of disc tissue resulting in either anti‐ or pro‐catabolic outcomes. This may help to explain the beneficial effects of motion‐based therapies as well as the destructive effect of traumatic levels of applied strain. © 2011 Orthopaedic Research Society Published by Wiley Periodicals, Inc. J Orthop Res 29: 1275–1283, 2011