Diabetes impairs stem cell and proangiogenic cell mobilization in humans

Abstract

Objective: Diabetes mellitus (DM) increases cardiovascular risk, at least in part, through shortage of vascular regenerative cells derived from the bone marrow (BM). In experimental models, DM causes morphological and functional BM alterations, but information on BM function in human DM is missing. Herein, we sought to assay mobilization of stem and proangiogenic cells in subjects with and without DM.

Research design and methods: In a prospective trial (NCT01102699), we tested BM responsiveness to 5 μg/kg human recombinant granulocyte colony-stimulating factor (hrG-CSF) in 24 individuals with DM (10 type 1 and 14 type 2) and 14 individuals without DM. Before and 24 h after hrG-CSF, we quantified circulating stem/progenitor cells and total and differential white blood cell counts. We also evaluated in vivo the proangiogenic capacity of peripheral blood mononuclear cells using the Matrigel plug assay.

Results: In response to hrG-CSF, levels of CD34(+) cells and other progenitor cell phenotypes increased in subjects without DM. Patients with DM had significantly impaired mobilization of CD34(+), CD133(+), and CD34(+)CD133(+) hematopoietic stem cells and CD133(+)KDR(+) endothelial progenitors, independently of potential confounders. The in vivo angiogenic capacity of peripheral blood mononuclear cells significantly increased after hrG-CSF in control subjects without DM, but not in patients with DM. DM was also associated with the inability to upregulate CD26/DPP-4 on CD34(+) cells, which is required for the mobilizing effect of granulocyte colony-stimulating factor.

Conclusions: Stem and proangiogenic cell mobilization in response to hrG-CSF is impaired in DM, possibly because of maladaptive CD26/DPP-4 regulation. These alterations may hamper tissue repair and favor the development of cardiovascular complications.

Figures

Figure 1

CD34+ cell mobilization after hrG-CSF. A: Absolute pre-G-CSF and post-G-CSF CD34+ cell count in all DM and all non–DM control (Ctrl) subjects (study primary end point). *P < 0.05 vs. baseline. Absolute pre-G-CSF and post-G-CSF CD34+ cell count in type 1 DM (T1D) compared with young control subjects (B) and in type 2 DM (T2D) compared with age-matched control subjects (C). *P < 0.05 vs. baseline. D: Changes in the absolute levels of circulating CD34+ cell counts. *P < 0.05 vs. DM.

Figure 2

Mobilization of other progenitor cell phenotypes after hrG-CSF. A–E: Absolute pre-G-CSF and post-G-CSF cell counts of circulating hematopoietic stem cells (CD133+, CD34+CD133+) and endothelial progenitor cells (CD34+KDR+, CD133+KDR+, and CD34+CD133+KDR+) in DM and non–DM patients. *P < 0.05 vs. baseline. F: Changes in the absolute levels of stem/progenitor cell phenotypes in DM and non–DM patients. *P < 0.05 vs. non–DM control subjects.

Figure 3

Effects of hrG-CSF on CD26/DPP-4 expression. Percentage CD26/DPP-4 expression on CD34+ cells was significantly (*P < 0.05) increased in non–DM control subjects, whereas it was reduced in DM patients.

Figure 4

In vivo proangiogenic cell function. Patients’ PBMCs before collected pre-G-CSF and post-G-CSF were embedded into Matrigel plugs and implanted in immunodeficient mice. Representative gross appearance of the plugs (scale bar = 1 cm) and histological sections (scale bar = 100 μm) are shown. In the bottom, quantification of hemoglobin content, a quantitative surrogate of perfusion, adjusted for mobilized monocytes is shown. *P < 0.05 vs. pre-G-CSF. †P < 0.05 vs. non–DM.