Mesenchymal stem cells (MSCs) are self-renewing, multipotent cells that could potentially be used to repair injured cartilage in diseases. The objetive was to analyze different sources of human MSCs to find a suitable alternative source for the isolation of MSCs with high chondrogenic potential.

Femoral bone marrow, adipose tissue from articular and subcutaneous locations (hip, knee, hand, ankle and elbow) were obtained from 35 patients who undewent different types of orthopedic surgery (21 women, mean age 69.83 ± 13.93 (range 38–91) years. Neoplasic and immunocompromised patients were refused. The Ethical Committee for Clinical Research of the Government of Aragón (CEICA) approved the study and all patients provided informed consent. Cells were conjugated wiith monoclonal antibodies. Cell fluorescence was evaluated by flow cytometry using a FACSCalibur flow cytometer and analysed using CellQuest software (Becton Dickinson). Chondrogenic differentiation of human MSCs from the various tissues at P1 and P3 was induced in a 30-day micropellet culture [Pittenger et al., 1999]. To evaluate the differentiation of cartilaginous pellet cultures, samples were fixed embedded in paraffin and cut into 5- υm-thick slices. The slices were treated with hematoxylin-eosin and safranin O (Sigma-Aldrich). Each sample was graded according to the Bern Histological Grading Scale [Grogan et al., 2006], which is a visual scale that incorporates three parameters indicative of cartilage quality: uniform and dark staining with safranin O, cell density or extent of matrix produced and cellular morphology (overall score 0–9). Stained sections were evaluated and graded by two different researchers under a BX41 dual viewer microscope or a Nikon TE2000-E inverted microscope with the NIS-Elements software. Statistics were calculated using bivariate analysis. Pearson's χ2 or Fisher's exact tests were used to compare the Bern Scores of various tissues. To evaluate the cell proliferation, surface marker expression and tissue type results, ANOVA or Kruskal-Wallis tests were used, depending on the data distribution. Results were considered to be significant when p was < 0.05.

MSCs from all tissues analysed had a fibroblastic morphology, but their rates of proliferation varied. Subcutaneous fat derived MSCs proliferated faster than bone marrow. MSCs from Hoffa fat, hip and knee subcutaneous proliferated slower than MSCs from elbow, ankle and hand subcutaneous. Flow cytometry: most of cells lacked expression of CD31, CD34, CD36, CD117 (c-kit), CD133/1 and HLA-DR. At same time 95% of cells expressed CD13, CD44, CD59, CD73, CD90, CD105, CD151 y CD166. Fenotype showed no differences in cells from different anatomic places. Cells from hip and knee subcutaneous showed a worst differentiation to hyaline cartilage. Hoffa fat cells showed high capacity in transforming to hyaline cartilage.

Cells from different anatomic places show different chondrogenic potential that has to be considered to choose the cells source.

NK cells participate in the control of infection and cell transformation but, on the other hand, they are involved in the pathology of different inflammatory disorders. Recent evidences suggest that inflammation is an important regulator of osteoarthritis, but the mechanism and cells responsible of inflammation maintenance are not well defined.

To understand the role of NK cells in osteoarthritis, we have performed a preliminary study to compare the phenotype and function of peripheral blood with synovial fluid NK cells from 49 patients with osteoarthritis undergoing total knee arthroplasty. A phenotype analysis of NK cells were carried out by flow cytometry using lineage surface marker. For the first time, the expression of granzyme A, granzyme B and perforin was also performed. Finally, cytotoxicity assays were carried out using previously isolated NK cells co-cultured with their natural target K562 cells.

The majority of NK cells from the synovial fluid were CD56brightCD16negative cells. Moreover, CD56brightCD16negative cells present in synovial fluid showed higher expression of granzyme A and low expression of granzyme B and perforin. In addition, and in contrast to NK cells isolated from the peripheral blood, synovial NK cells were not able to kill K562 cells.

Our results indicate that NK cells from the synovium of patients with osteoarthritis, which present an immunoregulatory non-cytotoxic phenotype, show a different to phenotype of NK cells from peripheral blood, preferably expressing granzyme A, a pro-inflammatory molecule which may contribute to the establishment of chronic articular inflammation in this type of patients.