Search

Background: NK cells have been long time considered as cytotoxic lymphocytes competent in killing virus-infected cells and tumors. However, NK cells may also play essential immuno-regulatory functions. In this context, the real existence of a defined NK subset with negative regulatory properties has been hypothesized but never clearly demonstrated., Methodology/principal Findings: Herein, we show the in vitro generation from human peripheral blood haematopoietic progenitors (PB-HP), of a novel subset of non-cytolytic NK cells displaying a mature phenotype and remarkable immuno-regulatory functions (NK-ireg). The main functional hallmark of these NK-ireg cells is represented by the surface expression/release of HLA-G, a major immunosuppressive molecule. In addition, NK-ireg cells secrete two powerful immuno-regulatory factors: IL-10 and IL-21. Through these factors, NK-ireg cells act as effectors of the down-regulation of the immune response: reconverting mature myeloid DC (mDC) into immature/tolerogenic DC, blocking cytolytic functions on conventional NK cells and inducing HLA-G membrane expression on PB-derived monocytes. The generation of "NK-ireg" cells is obtained, by default, in culture conditions favouring cell-to-cell contacts, and it is strictly dependent on reciprocal trans-presentation of membrane-bound IL-15 forms constitutively and selectively expressed by human CD34(+) PB-HP. Finally, a small subset of NKp46(+) HLA-G(+) IL-10(+) is detected within freshly isolated decidual NK cells, suggesting that these cells could represent an in vivo counterpart of the NK-ireg cells., Conclusions/significance: In conclusion, NK-ireg cells represent a novel truly differentiated non-cytolytic NK subset with a self-sustainable phenotype (CD56(+) CD16(+) NKp30(+) NKp44(+) NKp46(+) CD94(+) CD69(+) CCR7(+)) generated from specific pSTAT6(+) GATA3(+) precursors. NK-ireg cells could be employed to develop new immuno-suppressive strategies in autoimmune diseases, transplant rejection or graft versus host diseases. In addition, NK-ireg cells can be easily derived from peripheral blood of the patients and could constitute an autologous biotherapic tool to be used combined or in alternative to other immuno-regulatory cells.

In a case with secondary myelofibrosis occurring after essential thrombocythemia, cytogenetic analysis revealed an isolated translocation t(X;17)(q27;q22) in all cells. We found that a bacterial artificial chromosome (BAC) encompassing the breakpoint on chromosome 17 long arm contained only one gene, NOG. We therefore investigated the occurrence of this rare breakpoint in myeloproliferative disorders (MPDs). We identified three more patients with a 17q abnormality in MPDs: myelofibrosis with myeloid metaplasia (MMM); chronic myeloid leukemia positive for t(9;22)(q34;q11) with additional t(4;17)(p15;q22) at diagnosis; and myelofibrosis complicating polycythemia vera. All three cases exhibited a split of BACs containing NOG. The protein encoded by NOG, noggin, acts as an antagonist to bone morphogenetic secreted protein 2 and 4 (BMP2 and BMP4). A comparative analysis of gene expression on Agilent 22K oligonucleotide microarrays in purified CD34+ cells from the blood of MMM patients showed significant downregulation of BMPR2, BMPR1B, BMP2, and BMP8; upregulation of BMP3 and BMP10; and a trend to lower expression of NOG. Thus, given that expression and release of BMPs are important in the induction of osteosclerosis and angiogenic activity, the observed BMP deregulations could be triggered by potential NOG genetic alterations in the four cases here described, and may contribute to the myelofibrotic process characterized by bone marrow stromal reaction including collagen fibrosis, osteosclerosis, and angiogenesis.

The CD44 cell surface molecule has been shown to be the principal cell surface receptor for hyaluronan (or hyaluronic acid), a glycosaminoglycan component of marrow extracellular matrix. However, its affinity for hyaluronan is not constitutive, since it depends on the cell type, the stage of differentiation and on activation by external stimuli including certain anti-CD44 antibodies and phorbol esters. Except for a few lymphoid cell lines, hematopoietic cells do not spontaneously bind hyaluronan and initial studies reported that, contrary to lymphocytes, myeloid cells could not be activated to bind hyaluronan. Because CD44 plays an important role in the initial phases of hematopoiesis, as shown by experiments using blocking anti-CD44 monoclonal antibodies, its capacity to mediate adhesion of primitive myeloid cells has been investigated. It was found that CD44 could mediate spontaneous adhesion to hyaluronan of immature myeloid cell lines KG1, KG1a, and TF1, which serve as a model for hematopoietic progenitors. However, despite expressing high amounts of CD44, no more than 15% of bone marrow progenitors could adhere to hyaluronan. Recent experiments have shown that a very important feature of CD44 is its capacity to be rapidly activated by certain antibodies and cytokines (GM-CSF and KL) from a low affinity to a high affinity state for hyaluronan. These data shed light on striking similarities in the functional regulation of CD44 and of the two integrin receptors VLA-4 (a4b1), and VLA-5 (a5b1), which are also expressed on hematopoietic progenitors. The relevance of these data to the regulation of normal hematopoiesis and mobilization of CD34+ progenitors in the view of cell grafting is analyzed. In addition, we show that in idiopathic myelofibrosis, the amount of hyaluronan is markedly increased in the extracellular matrix from the myeloproliferative spleen. Considering that the production of cytokines is enhanced in this disease, we discuss whether CD44-hyaluronan interaction may have a role in the pathophysiology of this myeloproliferative syndrome.

The changes occurring in the hematopoietic extracellular matrix in an experimental myeloproliferative syndrome were explored by comparing the glycosaminoglycan (GAG) composition of normal mouse spleens and spleens infected with myeloproliferative sarcoma virus (MPSV). Large quantities of hyaluronate and of sulfated GAGs accumulated in the extracellular matrix of infected spleens, as shown by histoimmunoassay and alcian blue staining, respectively. The splenic GAGs were either labeled with 35S-sulfate injected in vivo or unlabeled. The spleens were fractionated to separate hematopoietic cells from the stromal component containing extracellular matrix material and fibroblasts, and the GAGs were extracted from each fraction. Specific degradative treatments and electrophoresis indicated that sulfated GAGs were mostly chondroitin sulfate and heparan sulfate. Three hours after in vivo injection of 35S-sulfate, the amount of 35S-GAGs was increased approximately fivefold per mg stromal proteins. The bulk of these 35S-GAGs (70%) was recovered in the stromal fraction. The higher amount of sulfated GAGs in leukemic spleen was due both to the presence of more producer cells (infected fibroblasts and hematopoietic cells) and to a stimulation of GAG synthesis per cell, as evidenced 35S-labeling in in vitro experiments. Chondroitin sulfate was the main sulfated GAG present in the culture medium of both hematopoietic and fibroblastic cells and in the pericellular material released by trypsin from fibroblastic cells. High amounts of chondroitin sulfate, which has a possible role in the detachment of hematopoietic cells from the stromal cells, may favour the release of hematopoietic cells from the spleen into the peripheral blood. Heparan sulfate was produced by fibroblastic cells and it was principally present in their pericellular material. Considering the capacity of heparan sulfate to retain cytokines, as demonstrated by others in vitro, large amounts of heparan sulfate may result in the retention of large amounts of the cytokines, which production is enhanced in the infected spleen. This phenomenon may contribute to promote the hematopoietic stem cell proliferation characteristic of the MPSV-induced myeloproliferative disease.

Receptors (R) are considered as allosteric enzymes whose action on metabolic chains is modulated through binding to the ligand. They play an essential role in the transduction of the multiple signals (e.g. interleukins or CSF) which intervene in the regulation of hematopoiesis. Their ordered interactions are necessary to regulate the growth and differentiation of normal hematopoietic precursors. This paper summarizes recent data concerning the structure-action relationship of growth-factor receptors in the signal transduction and alterations of growth-factor receptors which may play an important role in leukemic transformation. Some therapeutic modulations of growth-factors cascades are also discussed.

Autocrine growth is a process whereby a cell both secretes and responds to a growth factor. This paper describes the stepwise malignant progression of leukemic cells which has been demonstrated in many experimental models of autocrine leukemic growth. In contrast, autocrine growth has not been proven as a major physiopathological mechanism for the growth of leukemic cells in vivo in human myeloid and lymphocytic leukemias. Growth-factor independency of human leukemic cell lines may be due to clonal selection.

A subpopulation of granulocyte and macrophage precursors (GM-CFUc) differentiating in the agar colony technique of Bradley and Metcalf into mature granulocytes and macrophages, without the addition of granulocyte and macrophage colony stimulating factor (GM-CSF), can be detected in MPSV infected mice. These precursors were detected 5 days after virus infection, reaching a maximal concentration of 1/10,000 spleen or bone marrow cells, 25 days after viral infection. The number of the added GM-CSF independent GM-CFUc was linearly correlated with the number of seeded MPSV hematopoietic cells. No GM-CSF producing cells could be detected in the MPSV spleen using normal bone marrow GM-CFUc as responder cells. Study of the GM-CSF sensitivity of the GM precursors has demonstrated the existence of two GM-CFUc populations in the MPSV spleen: a) a GM-CSF dependent population with a GM-CSF sensitivity similar to that of normal GM-CFUc b) a GM-CFUc population which differentiated in the absence of detectable amount of GM-CSF and of which differentiation was not affected by the addition of progressive amounts of GM-CSF. A possible model explaining these results is proposed.