The stromal cell-derived factor-1 (SDF-1)/chemokine C-X-C receptor 4 (CXCR4) axis plays a critical role in homing, engraftment and retention of hematopoietic stem/progenitor cells. We previously demonstrated that expression of CD9 is a downstream signal of the SDF-1/CXCR4 axis, and that CD9 regulates short-term (20 hours) homing of cord blood (CB) CD34+ cells in the NOD/SCID mouse xenotransplantation model (Leung et al, Blood, 2011). Here, we provided further evidence that pretreatment of CB CD34+ cells with a CD9-neutralizing antibody significantly reduced their long-term (6 weeks) engraftment, as indicated by the presence of human CD45+ cells, in the recipient bone marrow and spleen by 70.9% (P = .0089) and 87.8% (P = .0179), respectively (n = 6). However, CD9 blockade did not bias specific lineage commitment, including the CD14+ monocytic, CD33+ myeloid, CD19+ B-lymphoid and CD34+ stem/progenitor cells (n = 4). We also observed an increase of the CD34+CD9+ subsets in the bone marrow (9.6-fold; P < .0001) and spleens (9.8-fold; P = .0014) of engrafted animals (n = 3-4). These data indicate that CD9 possesses important functions in regulating stem cell engraftment and its expression level on CD34+ cells is up-regulated in the target hematopoietic organs. Analysis of paired bone marrow (BM) and peripheral blood (PB) samples from healthy donors revealed a higher CD9 expression in BM-resident CD34+ cells (57.3% ± 8.1% CD9+ cells in BM vs. 29.3% ± 5.8% in PB; n = 5, P = 0.0478). Consistently, CD34+ cells in granulocyte colony-stimulating factor (G-CSF)-mobilized peripheral blood (MPB) expressed lower levels of CD9 (33.8% ± 3.0% CD9+ cells, n = 24), when compared with those in BM (56.4% ± 4.9% CD9+ cells, n = 8, P = 0.0025). In vitro exposure of MPB CD34+ cells to SDF-1 significantly enhanced CD9 expression (1.55-fold increase, n = 4, P = 0.0103), concomitant with a 75.2% reduction in the CD34+CXCR4+ subsets (P = 0.0118). Treatment of NOD/SCID chimeric mice with G-CSF increased the frequency of circulating CD45+ cells (3.4-fold) and CD34+ cells (3.3-fold), and substantially decreased the CD34+CD9+ subsets in the BM from 75.8% to 30.8%. Importantly, the decline in CD9 levels during G-CSF mobilization was also observed in the CD34+CD38-/low primitive stem cell subpopulation. Interestingly, in vitro treatment of BM CD34+ cells with G-CSF did not affect CD9 expression (n = 3), suggesting that a signaling intermediate is required for G-CSF-mediated CD9 down-regulation in vivo. Transwell migration assay revealed a significant enrichment of CD9- cells that were migrated towards a SDF-1 gradient (n = 4 for BM CD34+ cells, P = 0.0074; n = 7 for CB CD34+ cells, P = 0.0258), implicating that CD9 might negatively regulate stem cell motility. In contrast, pretreatment with the CD9-neutralizing antibody inhibited adhesion of CD34+ cells to the osteoblastic cell line Saos-2 by 33.5% (n = 2). Our results collectively suggest a previously unrecognized role of CD9 in stem cell retention by dual regulation of cell motility and adhesion, and reveal a dynamic regulation of CD9 expression in the BM microenvironment, which might represent an important event in controlling stem cell homing and mobilization.

Disclosures:

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution