Background Calcific tendinopathy is one of the most frequent causes of shoulder pain. Calcific deposits lead to chronic discomfort in daily and professional activity. These deposits are composed of carbonated apatite. Although the disease is frequent, its origin stays still largely unknown. Molecular and cellular mechanisms involved in this pathological mineralization process are not clearly identified. Objectives The objective of the study was to analyze calcified tendinous samples to understand the organization of the deposits and to characterize the cells potentially involved in their formation. Methods Samples were collected from cadaveric subjects. Ultrasound was first used to detect calcified tendons. Then, tendons were collected and fixed in formalin 4% during 48h. They were first analyzed with micro-CT to know the distribution of the calcific deposits. Samples were then decalcified in EDTA, dehydrated and embedded in paraffin. Some samples were not decalcified to allow a better characterization of the calcific deposits. Several histological staining were performed: hematoxylin and eosin (HE), Safranin O/Fast Green (SO/FG) and Von Kossa (no decalcified samples). Immunohistochemistry using anti-Runx2, anti-Sox9, anti-Collagen II and anti-caspase III antibodies has been performed to characterize the cells and tissue around the calcifications. Results Six samples were collected (1 normal and 5 calcified). On HE staining, three different histological patterns were observed. Little calcifications disseminated between tendinous fibers (N=2), voluminous ones encapsulated by a fibrous tissue (N=2) and in one sample an intra-tendinous osseous metaplasia. In the fibrous peripheral area of larger calcifications, we observed cells with round nuclei, different from tenocytes. These cells expressed Runx2 and Sox9 suggesting a chondrocyte phenotype. On SO/FG staining, this peripheral area presented a red coloration (proteoglycan specific) as the fibrocartilage at the tendon attachment. However, collagen II clearly present in the fibrocartilage was not present in these areas. As pathological calcification in cartilage can be associated with chondrocytes apoptosis, we sought for anti-Caspase III expression in the cells of the peripheral area. None of the chondrocyte-like cells located around the larger calcifications expressed Caspase III. Finally, one sample had an osseous metaplasia within the tendon with Runx2 positive cells. Conclusions Histological analyses of whole calcified tendon tissues showed three different patterns of calcific deposits. We can hypothesize that these patterns correspond to different stages of the disease. Chondrocyte-like cells were observed around larger deposits and could be involved in the mineralization process. Interestingly, they differ from the cells of the fibrocartilage as they did not express collagen II. Further analyses are necessary to characterize their phenotype and understand the steps leading to these deposits within the tendon. Disclosure of Interest None declared
