RNA-based bone histomorphometry: method and its application to explaining postpubertal bone gain in a G610C mouse model of osteogenesis imperfecta

Bone histomorphometry is a well-established approach to assessing skeletal pathology, providing a standard evaluation of the cellular components, architecture, mineralization, and growth of bone tissue. However, it depends in part on the subjective interpretation of cellular morphology by an expert, which introduces bias. In addition, diseases like osteogenesis imperfecta (OI) and fibrous dysplasia are accompanied by changes in the morphology and function of skeletal tissue and cells, hindering consistent evaluation of some morphometric parameters and interpretation of the results. For instance, traditional histomorphometry combined with collagen turnover markers suggested that reduced bone formation in classical OI is accompanied by increased bone resorption. In contrast, the well-documented postpubertal reduction in fractures would be easier to explain by reduced bone resorption after puberty, highlighting the need for less ambiguous measurements. Here we propose an approach to histomorphometry based on in situ mRNA hybridization, which uses Col1a1 as osteoblast and Ctsk as osteoclast markers. This approach can be fully automated and eliminates subjective identification of bone surface cells. We validate these markers based on the expression of Bglap, Ibsp, and Acp5. Comparison with traditional histological and tartrate-resistant acid phosphatase staining of the same sections suggests that mRNA-based analysis is more reliable. Unlike inconclusive traditional histomorphometry of mice with alpha 2(I)-Gly610 to Cys substitution in the collagen triple helix, mRNA-based measurements reveal reduced osteoclastogenesis in 11-wk-old animals consistent with the postpubertal catch-up osteogenesis observed by microCT. We optimize the technique for cryosections of mineralized bone and sections of paraffin-embedded decalcified tissue, simplifying and broadening its applications. We illustrate the application of the mRNA-based approach to human samples using the example of a McCune-Albright syndrome patient. By eliminating confounding effects of altered cellular morphology and the need for subjective morphological evaluation, this approach may provide a more reproducible and accessible evaluation of bone pathology. Maintenance of healthy bone depends on the cooperative action of bone-building cells (osteoblasts) and bone-resorbing cells (osteoclasts). Analysis of histological sections for bone surface coverage by osteoblasts and osteoclasts is thus essential in skeletal biology and pathology studies. This bone histomorphometry is, however, known to be affected by the misidentification of some osteoblasts and osteoclasts due to ambiguous cell morphology and the lack of sufficiently specific staining in traditional osteoclast assays. We propose staining histological sections for RNA transcripts of key marker genes to unambiguously identify osteoblasts and osteoclasts. We select the best marker genes, optimize the assay, and demonstrate its improved accuracy and reproducibility compared to the traditional analysis in a mouse model of osteogenesis imperfecta, a heritable disorder causing bone fragility. Using this model, we illustrate how the increased accuracy of the RNA-based histomorphometry may explain the puzzling bone improvement after puberty in the animals with the disorder. We demonstrate that the same approach can be used for bone histomorphometry in humans and propose its evaluation for implementation in clinical practice. Graphical Abstract