Circulating endothelial progenitor cells are not affected by acute systemic inflammation

Abstract

Vascular injury causes acute systemic inflammation and mobilizes endothelial progenitor cells (EPCs) and endothelial cell (EC) colony-forming units (EC-CFUs). Whether such mobilization occurs as part of a nonspecific acute phase response or is a phenomenon specific to vascular injury remains unclear. We aimed to determine the effect of acute systemic inflammation on EPCs and EC-CFU mobilization in the absence of vascular injury. Salmonella typhus vaccination was used as a model of acute systemic inflammation. In a double-blind randomized crossover study, 12 healthy volunteers received S. typhus vaccination or placebo. Phenotypic EPC populations enumerated by flow cytometry [CD34(+)VEGF receptor (VEGF)R-2(+)CD133(+), CD14(+)VEGFR-2(+)Tie2(+), CD45(-)CD34(+), as a surrogate for late outgrowth EPCs, and CD34(+)CXCR-4(+)], EC-CFUs, and serum cytokine concentrations (high sensitivity C-reactive protein, IL-6, and stromal-derived factor-1) were quantified during the first 7 days. Vaccination increased circulating leukocyte (9.8 + or - 0.6 vs. 5.1 + or - 0.2 x 10(9) cells/l, P < 0.0001), serum IL-6 [0.95 (0-1.7) vs. 0 (0-0) ng/l, P = 0.016], and VEGF-A [60 (45-94) vs. 43 (21-64) pg/l, P = 0.006] concentrations at 6 h and serum high sensitivity C-reactive protein at 24 h [2.7 (1.4-3.6) vs. 0.4 (0.2-0.8) mg/l, P = 0.037]. Vaccination caused a 56.7 + or - 7.6% increase in CD14(+) cells at 6 h (P < 0.001) and a 22.4 + or - 6.9% increase in CD34(+) cells at 7 days (P = 0.04). EC-CFUs, putative vascular progenitors, and the serum stromal-derived factor-1 concentration were unaffected throughout the study period (P > 0.05 for all). In conclusion, acute systemic inflammation causes nonspecific mobilization of hematopoietic progenitor cells, although it does not selectively mobilize putative vascular progenitors. We suggest that systemic inflammation is not the primary stimulus for EPC mobilization after acute vascular injury.

Fig. 1.

Flow cytometric analysis. Leukocytes were identified on the basis of their characteristic forward (FSC) and side scatter (SSC) profile (A). Within this population, CD34⁺-FITC, CD133⁺-PE, CD45⁺-PercP, and VEGF receptor (VEGFR)-2⁺-APC events were identified (B). Using Boolean principles (“and,” “not,” “or,” etc.), double- or triple-positive events were determined. For CD14⁺ populations, coexpression with Tie-2-APC and or VEGFR-2-PE was determined using quadrant analysis (C). The fluorescence minus one technique was used, and gates were set on single or unstained stained negative controls where appropriate. Negative controls are shown on the left and stained samples are shown on the right in B and C.

Fig. 2.

Serum cytokines in response to Salmonella typhus vaccination. Vaccination resulted in an increase in serum IL-6 (P = 0.016) and VEGF-A (P = 0.009) concentrations at 6 h and a later increase in high sensitivity C-reactive protein (hs-CRP) concentrations (P < 0.001) at 24 h. Serum stromal-derived factor (SDF)-1 concentrations were unaffected by either placebo or vaccination. Data are presented as median changes from baseline ± interquartile ranges.

Fig. 3.

Circadian variation of CD34⁺ progenitor cells. Circadian variation of CD34 populations was demonstrated after placebo, with significant reductions in CD34⁺ (P = 0.033), CD34⁺VEGFR-2⁺ (P = 0.001), and CD34⁺CD133⁺VEGFR-2⁺ (P = 0.037) cells by midafternoon.

Fig. 4.

Effect of vaccination on circulating CD34⁺ populations and endothelial cell (EC) colony-forming units (EC-CFUs). After vaccination, the fall in CD34⁺ populations from baseline at 6 h was attenuated (A–C), and a significant increase in CD34⁺ cells occurred at 1 wk (P = 0.04; A). However, CD45⁻CD34⁺, CD34⁺VEGFR-2⁺, and CD34⁺CD133⁺VEGFR-2⁺ cells were not altered by vaccination (P = 0.4, 0.32, and 0.39, respectively; C–E). EC-CFUs were similarly unaffected (F). Data are presented as absolute mean changes from baseline ± SE in A–E and as absolute median changes from baseline ± interquartile ranges in F.

Fig. 5.

Effect of vaccination on circulating CD14⁺ populations. Vaccination resulted in a significant mobilisation of CD14⁺ cells compared with placebo (P < 0.001; A). CD14⁺ subpopulations bearing Tie-2 and/or VEGFR-2 were not significantly different from placebo throughout the study period (P > 0.2; B–D). Data are presented as absolute mean changes from baseline ± SE.

Fig. 6.

EC-CFU characterization. A: a typical EC-CFU shown at ×20 magnification. Colonies displayed the classical, although nonspecific, characteristics of ECs: uptake of Alexa fluor 546-conjugated acetylated LDL (B) and FITC-conjugated Ulex europaeus binding (C). Colonies also expressed endothelial nitric oxide synthase (D), an enzyme responsible for the production of nitric oxide by ECs in response to sheer stress, Tie-2 (E), a tyrosine kinase receptor expressed on mature ECs necessary for normal vascular development, and CD146 (F), a transmembrane glycoprotein responsible for endothelial intracellular adhesion. CD105, or endoglin, a constituent of the transforming growth factor-β receptor, is a marker of the activated endothelium thought to be limited to proliferating cells and was widely expressed throughout the colonies (G). The hematopoietic nature of the colonies was evident by their widespread expression of CD45 (H) and, most intensely of all, the macrophage marker CD68 (I). Expression was at a magnitude similar to positive controls (data not shown). Magnification: ×40 in B–I. Colonies in E–I were stained with primary antibodies followed by the relevant secondary antibody-fluorochrome conjugate. Nuclei were counterstained with 4′,6-diamidino-2-phenylindole (blue).