Splenic extramedullary hematopoiesis is a major clinico-pathological feature of patients with myelofibrosis. As in the bone marrow (BM), hematopoiesis in the spleen occurs thank to the interplay of hematopoietic progenitor cells with the microenvironment, which provides the regulatory mechanism for their differentiation, proliferation and trafficking. Among other components, such as vessels and extra-cellular matrix proteins, this microenvironment encompasses different types of accessory cells, including mesenchymal stromal cells (MSCs). We have recently reported that MSCs from the BM of patients with myelofibrosis harbor genetics abnormalities and display an altered functional activity, suggesting that a primary MSC defect may either lead to or favor the pathogenesis of the disease. Here, we describe the phenotypical, functional, and genetic profile of MSCs isolated from the spleen of 23 patients with myelofibrosis, who underwent splenectomy for anemia and/or for excessive size of the spleen, and compare them to splenic mesenchymal stromal cells (s-MSCs) from 7 healthy subjects (HSs) who were splenectomized following traumatic lesion. The study was approved by the institutional review board of IRCSS Policlinico San Matteo Foundation; patients and HSs gave written informed consent for participating to the study. Mononuclear cells (MNCs) were obtained by dissociation of small spleen fragments by means of the GentleMacs Dissociator device (Miltenyi Biotech, Germany), and s-MSCs were isolated and expanded according to the standard procedures used for BM-MSCs. S-MSCs were obtained in 9/23 patients and in 3/7 HSs and displayed no significant differences for morphology and differentiation ability into adipocytic and osteoblastic lineages. However, the clonogenic efficiency of s-MSCs from patients with myelofibrosis was statistically higher than that of HSs (0.07 colonies/106 MNCs, range 0.03-0.01, vs 0.03/106 MNCs, range 0.03-0.04, respectively; p=0.048), whereas doubling time and time to senescence were not statistically different. Flow cytometric assessment of standard surface antigens (CD13, CD14, CD34, CD45, CD73, CD90, CD105) confirmed the mesenchymal nature of the cells grown in the cultures, and was similar between patients’ and HSs’ s-MSCs. When nestin expression was determined, no significant differences in the frequency of MSCs expressing this antigen was observed; however, nestin Mean Fluorescence Intensity (MFI) of patients’ s-MSCs was significantly lower than that of s-MSCs from HSs (22, range 6-45, vs 97, range 65-100, respectively; p=0.035). Patients’ s-MSCs also displayed a reduced capacity to sustain long term hematopoiesis in vitro in a classical Long Term Culture-Initiating Cell assay. However, when normal cord blood-derived CD34+ cells were co-cultured onto patients’ s-MSCs in a transwell system for 13 days, the output of CD41+ megakaryocytic cells increased with respect to culture where CD34+ cells were plated onto HSs’ s-MSCs [21,5% vs 14,2% w/o recombinant human thrombopoietin (rhTPO), respectively, p=0,043; 60,2% vs 33,6% with rhTPO, respectively, p=0,01] at detriment of CD33+ cells (41,5% vs 48,6% w/o rhTPO, respectively, p=0,049; 10,4% vs 29,4% with rhTPO, respectively, p=0,012]. Finally, an abnormal karyotype [46XXt(5;17)(4-12)] was detected in 1 out of 18 metaphases of 1 out of 3 patient s-MSCs, while a normal karyotype was always observed in 2 out of 2 HSs’ s-MSC. This extensive characterization of s-MSCs shows that s-MSCs of patients with myelofibrosis display functional and genetic abnormalities compared to those isolated from HSs. The low level of nestin expression suggests that the hematopoietic niche of the spleen of patients with myelofibrosis can be defective and responsible for the increased trafficking of CD34+ cells that is observed in these patients, whereas the increased differentiation into the megakaryocytic lineage indicates a role of the splenic niche in leading hematopoiesis toward a pathological profile. All together, our data suggest that s-MSCs play a role in the pathogenesis of myelofibrosis and could be, therefore, a potential target for the treatment of the disease. DisclosuresNo relevant conflicts of interest to declare.
Read full abstract