Prostate cancer is frequently associated with bone metastases, which are in fact the main cause of morbidity and mortality for this tumor. To better investigate this process, animal models of bone and bone marrow metastases need to be developed. However, experimental prostate cancer bone metastases are difficult to be obtained in vivo, and some typical clinical patterns remain irreproducible. In this study, we injected PC3 prostate cancer cells into the left cardiac ventricle of nude mice, thus reproducing the basic biological phenomenon of tumor cell spreading in the arterious blood stream, and compared the outcome with direct injection of cells in the bone marrow cavity of the tibia. Mice were monitored by X-ray analysis. After 40 days, 100% of intratibially-injected and 64% of heart-injected mice revealed osteolytic lesions. The heart injection was then used to select PC3 cell subpopulations by serial inoculation and recovery from bone metastasis. One of the resulting cell populations, obtained by a second round of selection and denominated PCb2, showed a more invasive phenotype compared to parental PC3, both in vitro and in vivo. Although PCb2 cells had a growth rate comparable to that of PC3 cells, they generated a higher number of bone lesions in nude mice and crossed more easily the Matrigel barrier in vitro in the presence of several chemoattractants. This phenomenon was partially due to an increased capacity to adhere to laminin and to release MMP-2 at higher level relative to the original PC3 cells. This study demonstrates that heart injection of prostate cancer cells in nude mice may represent a good experimental model to investigate the pathophysiology of bone and bone marrow metastases in vivo.

Related Articles