CD26 Expression on Neutrophils Is Critical for G-CSF Mediated Mobilization of Hematopoietic Stem and Progenitor Cells.

Abstract

Peripheral blood stem cells (PBSC) mobilized by G-CSF are the primary source of hematopoietic stem (HSC) and progenitor cells (HPC) for hematopoietic transplantation, however, the detailed mechanism of G-CSF-mediated HS/PC mobilization is not completely understood. Recent studies (Christopherson et al. Exp Hematol , 31:1126, 2003) suggest that CD26 is essential for mobilization by G-CSF, but it is not clear whether CD26 on HS/PC mediate this event. CD26 is a membrane-bound peptidase that degrades the N-terminus of chemokines including CXCL12/SDF-1. We found that the G-CSF-mobilized PBSC population in mice contains equal numbers of CD26− and CD26+ KSL (58%±14 vs. 42%±12) and KL cells (65%±12 vs 35%±7), suggesting that HS/PC CD26 expression may not be required for their mobilization. Analysis of CD26 expression on KSL and KL cells after G-CSF demonstrate that the total number of bone marrow (BM) CD26+ KSL and KL cells as well as MFI of CD26 on these cells remained unchanged over the 4 days of G-CSF treatment. In contrast, CD26− KSL and KL cells increased by 87%±10 and 60%±7 respectively. The CD26 inhibitor diprotin A reduced the number of CFU mobilized by 45±10% compared to G-CSF alone as previously described. These observations support the hypothesis that CD26 expression on a cell type other than HS/PC is required for G-CSF mobilization. Since we previously demonstrated a role for neutrophils (PMN) in G-CSF-mediated mobilization and PMN also express CD26, we evaluated the requirement for PMN CD26 in G-CSF mediated mobilization. After 2 and 4 days of G-CSF treatment, CD26 expressing PMN (Gr-1+, CD26+) increased in BM by 50±6% and 70±9% respectively. Depletion of CD26 + and - PMN by anti-Gr-1 Ab during G-CSF treatment resulted in loss of HSC mobilization in spite of equivalent numbers of CD26 expressing KSL and KL cells in BM compared to control mice. To further determine whether HS/PC extrinsic CD26 expression is required for mobilization by G-CSF, a series of mixed BM chimeric mice were generated by transplanting BM cells from CD26 knockout (−/−) (CD45.1) and wild-type (+/+) (CD45.2) congenic mice at ratios of 1:1, 2.5:1 and 5:1(−/−: +/+) into irradiated CD45.2 mice. We hypothesized that if HS/PC CD26 expression is required for their mobilization, then treatment with G-CSF would mobilize only +/+ HS/PC. However at 2 months post transplant, we observed equal mobilization of both CD26 +/+ and −/− KSL cells (1:1 chimeras- 3.8 vs 3.6-fold; 2:1 chimeras- 3.2 vs 3.0-fold respectively) into the blood after G-CSF treatment. In these chimeric mice, nucleated marrow cellularity and peripheral blood counts were fully restored and the degree of mixed chimerism in BM and peripheral blood (PB) was consistent with the cell ratios transplanted, with ∼50 and 35% of total PMN expressing CD26+. In the 5:1 chimeric mice, mobilization of both CD26 −/− and +/+ KSL cells was significantly reduced (1.5 vs 1.4-fold, respectively), being equivalent to that observed in CD26 −/− mice. Reduced mobilization was associated with normal numbers of PMN however <10% of PMN expressed CD26. In conclusion, this study suggests that CD26 expressing PMN generate trans-acting signals required for mobilization of CD26 −/− as well as CD26+ HS/PC by G-CSF.