CLL Cell-Derived Exosomes Promote Bone Marrow Fibrosis and Inhibit Normal Hematopoietic Colony Proliferation

Whereas low grade bone marrow (BM) reticulin fibrosis is detected in most patients with CLL, high grade BM fibrosis is associated with anemia, thrombocytopenia and poor prognosis. Although this observation was reported several years ago, the mechanism of BM fibrosis in CLL is still unknown. We have recently found that CD14+ monocyte-derived fibrocytes, known to participate in the initiation of tissue fibrosis in the skin, lung, kidney, liver, and heart, induce BM fibrosis in patients with primary myelofibrosis. Therefore, we wondered whether fibrocytes induce BM fibrosis in CLL. Consistent with our hypothesis, we found increased fibrocytes numbers in CLL patients' BM biopsies as assessed by immunohistochemistry analysis. Because fibrocytes are derived from CD14+ monocytes, thought to derive from the normal hematopoietic clone, we wondered what induces overproduction of fibrocytes in the BM of patients with CLL.

Exosomes are 60 -100 nm vesicles released by mammalian cells. Exosomes harbor various intracellular components including transcriptionally active mRNA, non-coding RNAs and proteins. Numerous studies demonstrated that tumor-derived exosomes are stable inter-cellular transport vehicles that deliver tumor cell cargo into other cells. Therefore we wondered whether exosomes affect the generation of BM fibrosis in CLL. In light of recent studies showing that bystander cells engulf CLL-derived exosomes which often alter their function, we hypothesized that CD14+ monocytes engulf CLL-derived exosomes that increase the capacity of these monocytes to differentiate into reticulin producing fibrocytes. To test this hypothesis, we incubated peripheral blood CLL cells in culture medium containing exosome-depleted fetal bovine serum for 2 days and harvested extracellular nanoparticles with CD63-coated beads. By using flow cytometry we found that the CD63+ nanoparticles co-express CD63, CD5, and CD19, and by electron microscopy we confirmed that they were typical 60 to 100 nm cup-shaped exosomes co-expressing CD5 and CD19 as assessed by immuno-gold. To test whether normal monocytes engulf CLL-derived exosomes we incubated CLL-derived exosomes stained with a green FM 1-43 dye, with healthy donor-derived CD14+ monocytes. After 2 hours, excess exosomes were washed out and the monocytes were counterstained with Evans blue. Confocal microscopy analysis showed that these monocytes harbor a significant number of CLL-derived exosomes. We also cultured these monocytes under conditions that favor fibrocyte differentiation and found that normal CD14+ monocytes incubated with CLL-derived exosomes, but not with CLL patients' filtered plasma, express significantly higher levels of CD14 surface antigen than untreated monocytes and form fibrocytes at a 3.6 higher rate than normal monocytes, similar to the level of CLL-derived monocytes. Because we previously found that normal fibrocytes stimulate hematopoietic colony proliferation, we wondered what would be the function of CLL-derived fibrocytes or of fibrocytes generated from normal CD14+ monocytes incubated with CLL exosomes. Compared to untreated controls, the number of colony-forming unit granulocyte-macrophage (CFU-GM) and burst-forming unit-erythroid (BFU-E) colonies was markedly increased when normal BM cells were cultured on top of normal fibrocytes. In contrast, the number of CFU-GM and BFU-E colonies was significantly lower when normal BM cells were cultured on top of CLL-derived fibrocytes or fibrocytes derived from normal CD14+ cells that were incubated with CLL-derived fibrocytes. Furthermore, CLL-derived exosomes, but not normal low-density cell-derived exosomes, inhibited CFU-GM and BFU-E colony growth in a dose-dependent manner, suggesting that CLL-derived exosomes directly inhibit normal myeloid progenitor proliferation.

In conclusion: CLL-derived exosomes reprogram monocytes to generate an increased number of reticulin-producing fibrocytes and directly and indirectly inhibit myeloid progenitor proliferation. Taken together our data show how exosomes contribute to hematopoietic dysfunction in CLL and provide a novel perspective on how CLL cells actively modify their microenvironment.