PubMed1994
Calcitonin inhibits bone resorption through a direct action on the osteoclast. We report a quantitative analysis of bone resorption by disaggregated rat osteoclasts. We then used our findings to develop a formal bioassay for calcitonin. Osteoclasts were mechanically disaggregated from neonatal rat long bones and dispersed at low densities on slices of devitalized bovine cortical bone. The resulting areas of bone excavation were quantified to micrometric precision by scanning electron microscopy together with computer-assisted image analysis. These findings were correlated with the volumes of bone resorption in the same slices measured by confocal scanning microscopy for the first time. The total planar areas of bone resorption per slice correlated linearly (r = 0.78) with the confocal microscopic measurements of total volume resorbed, provided that volume was expressed to its two-thirds power. The latter transformation resulted in representations of the determined areas ([length]2) and volumes ([length]3) which were dimensionally consistent. These findings thus demonstrate that osteoclastic bone excavations show a consistent relationship between area and volume and that assessments of the area of excavations accordingly provide an empirical representation of the volume of bone resorbed. Furthermore, in view of the skewed nature of the distributions of area measurements, we assessed the effect of transforming the response variable to derive a metameter, (planar area of resorption)1/2. Such transformed data points, which expressed the data in the dimensions of [length], were more normally distributed than the raw data points and had more stable variances over a wider concentration range. We accordingly determined relative potencies using parallel line analyses on the transformed data. The latter offered a consistent correlation to the volume measurements when these were also converted to dimensions of [length] (r = 0.805). It was confirmed that the inhibition of bone resorption by calcitonins from various species, namely, pig, salmon and eel, was quantitatively dependent upon concentration of the respective peptides. The resulting assay was also found to be sufficiently sensitive to measure picomolar peptide concentrations with a precision, lambda (standard deviation/slope), ranging between 0.3 and 0.8. Finally, we identified factors affecting assay precision and sensitivity.