The mechanism of G-CSF-induced hematopoietic progenitor/stem cell (HPC/HSC) mobilization from bone marrow microenvironment to peripheral blood has been related to local production by neutrophils of proteases, including elastase, cathepsin G and metalloproteinases (MMPs). We previously showed that in vitro migration of hematopoietic cells across a layer of human marrow stromal cells (MSC) is induced when stimulated by G-CSF. In the present study, we investigated the role of MMPs produced by MSC used for the trans-stromal migration of MO7e line and marrow CD34+ cells. Normal human MSC were previously grown to confluence on Transwell® filters (pore diameter 5-μm) and then stimulated or not by IL-1 (15 U/mL/Day) or G-CSF (15 ng/mL/Day) for three consecutive days. In a second step, MO7e or marrow CD34+ cells, put in the upper chamber, were allowed to migrate through the layers (4 hrs, SDF-1 100 ng/mL in the bottom chamber). The contribution of MMPs was evaluated in this in vitro trans-stromal migration assay using specific blocking monoclonal antibodies (mAb), anti-MMP-1, anti-MMP-2 and anti-MMP-9 (each at 5 μg/mL). The amounts of antigenic MMP-2 and MMP-9 produced by MSC were evaluated in cell supernatants by enzyme-linked immunosorbent assay (ELISA) and gelatinolytic activity of MMPs by zymography. mRNA expression of MMPs in MSC was analyzed by reverse transcriptase-polymerase chain reaction (RT-PCR). None of the MSC layers used contained detectable hematopoietic cells since they were all CD45-. Migration of MO7e cells across unstimulated or G-CSF-stimulated stromal cells was significantly inhibited by specific mAb. Percentages of inhibition were 25+/−5 % and 54+/−9 % for anti-MMP-2 mAb, and 18+/−6 % and 53+/−9 % for anti-MMP-9 mAb, respectively. Using the combination of anti MMP-2 and anti-MMP-9, inhibition reached 71+/−24 % across stimulated layers. In contrast, anti-MMP-1 mAb did not show any significant inhibition. Migration of CD34+ across G-CSF-stimulated stromal cells was similarly inhibited with percentages of 38+/−18 % for anti-MMP- 2 mAb, and 34+/−11 % for anti-MMP-9 mAb, while migration was not inhibited again, using anti-MMP-1 mAb. In parallel, we demonstrated that MSC did express G-CSF receptor by RT-PCR. Production of MMPs by unstimulated and G-CSF-stimulated MSC was detected by ELISA, zymography and RT-PCR for MMP-2, and by zymography and RT-PCR for MMP-9. In conclusion, G-CSF-induction of HPC trans-stromal migration involves MMP-2 and MMP-9 but not MMP-1. These findings suggest that microenvironment, in the absence of neutrophils, could have a significant role in G-CSF mobilization of HPC/HSC. A local production of MMP-2 and MMP-9 by marrow stromal cells could be critical in the egress of HSC out of the hematopoietic niche.

Author notes

Corresponding author

Sign in via your Institution