Bone Sialoprotein Shows Enhanced Expression in Early, High-Proliferation Stages of Three-Dimensional Spheroid Cell Cultures of Breast Cancer Cell Line MDA-MB-231

Normally, bone sialoprotein (BSP) is an important contributor to bone micro-calcification. However, it is also highly expressed in bone-metastatic malignancies, including prostate, lung, and breast cancer. In these disorders, BSP correlates with poor prognosis. Its expression in triple-negative breast cancer cells is enhanced by the transcription factor RUNX2, and both, BSP and RUNX2 are under control of IGF-1 and TGFβ1. Knockdown of BSP or its inactivation by specific antibodies were found to reduce the metastatic potential of MDA-MB-231 triple-negative breast cancer cells in xenografts. While the role of BSP in bone metastasis was studied using such models, valid test systems to investigate BSP biology have been lacking since this protein is expressed at very low levels in classical 2D cell cultures and the frequently used breast cancer cell line MDA-MB-231 is difficult to grow in 3D. Here, we have developed a long-term 3D spheroid culture model using MDA-MB-231 cells in a sandwich approach using cell embedding between a non-adherent surface and basement membrane extracts. This allowed consistent growth of spheroids for more than 21 days. Also, co-culturing of MDA-MB-231 with CCD-1137Sk fibroblasts yielded stably growing spheroids, suggesting the importance of extracellular matrix (ECM) in this process. In addition, we have set up a novel and simple open source analysis tool to characterize protein expression in 2D cultures and spheroids by immunofluorescence. Using this approach in combination with Western blot analysis, the expression profile of BSP was analyzed. BSP was enriched at the rims of spheroids, both in mono- and co-cultures and its abundance in general correlated with that of TGFβ1 under different conditions, including spheroid maturation, cytostatic treatment, and fibroblast co-culture. Conversely, correlation of IGF-1 and BSP was limited to mono-culture time course profiles. In conclusion, we present novel tools to study the regulation of gene expression in combination with cell proliferation and apoptosis in a long-term 3D model of breast cancer and find dynamic abundance profiles of the metastasis-relevant protein BSP and its regulators.