Absence of a relationship between immunophenotypic and colony enumeration analysis of endothelial progenitor cells in clinical haematopoietic cell sources

Abstract

Background: The discovery of adult endothelial progenitor cells (EPC) offers potential for vascular regenerative therapies. The expression of CD34 and VEGFR2 by EPC indicates a close relationship with haematopoietic progenitor cells (HPC), and HPC-rich sources have been used to treat cardiac and limb ischaemias with apparent clinical benefit. However, the laboratory characterisation of the vasculogenic capability of potential or actual therapeutic cell autograft sources is uncertain since the description of EPC remains elusive. Various definitions of EPC based on phenotype and more recently on colony formation (CFU-EPC) have been proposed.

Methods: We determined EPC as defined by proposed phenotype definitions (flow cytometry) and by CFU-EPC in HPC-rich sources: bone marrow (BM); cord blood (CB); and G-CSF-mobilised peripheral blood (mPB), and in HPC-poor normal peripheral blood (nPB).

Results: As expected, the highest numbers of cells expressing the HPC markers CD34 or CD133 were found in mPB and least in nPB. The proportions of CD34+ cells co-expressing CD133 is of the order mPB>CB>BM approximately nPB. CD34+ cells co-expressing VEGFR2 were also most frequent in mPB. In contrast, CFU-EPC were virtually absent in mPB and were most readily detected in nPB, the source lowest in HPC.

Conclusion: HPC sources differ in their content of putative EPC. Normal peripheral blood, poor in HPC and in HPC-related phenotypically defined EPC, is the richest source of CFU-EPC, suggesting no direct relationship between the proposed EPC immunophenotypes and CFU-EPC potential. It is not apparent whether either of these EPC measurements, or any, is an appropriate indicator of the therapeutic vasculogenic potential of autologous HSC sources.

Figure 1

Expression of CD34 or CD133 markers in haematopoietic stem cell clinical transplant sources and normal blood. Total numbers of cells expressing CD34 (a) or CD133 (b) in the different sources tested. Sources were normal (non-mobilised) peripheral blood (nPB), bone marrow (BM), umbilical cord blood (CB) and G-CSF-mobilised peripheral blood samples (mPBP) from patients for autologous grafts.

Figure 2

Relative proportions of CD34+ and/or 133+ cells in different sources. Percentages (of their summed populations) of CD34⁺ cells negative for CD133 expression (blue), CD133⁺ cells negative for CD34 expression (yellow) and cells co-expressing CD133 and CD34 markers (green) were compared between various haematopoietic stem cell clinical transplant sources and normal blood. Normal (non-mobilised) peripheral blood (nPB), bone marrow (BM), umbilical cord blood (CB) and G-CSF-mobilised peripheral blood samples (mPBp, autologous patients; or mPBd, allogeneic donors).

Figure 3

Number and percentage of putative EPC characterised by different phenotypic definitions in haematopoietic stem cell clinical transplant sources and normal blood. Number 3a) and percentage 3d) of CD34+ cells co-expressing VEGFR2 in the different sources tested.; Number 3b) and percentage 3e) of CD133+ cells co-expressing VEGFR2 in the different sources tested and Number 3c) and percentage 3f) of CD34⁺CD133⁺ cells co-expressing VEGFR2 in the different sources tested. Note different Y axis scales between figures.

Figure 4

Endothelial progenitor cell colony assay (CFU-EPC) in haematopoietic stem cell clinical transplant sources and normal blood. Endothelial progenitor cell colonies (CFU-EPC) per 10⁶ cells plated in the different sources tested. Normal peripheral blood (nPB)(n = 15), bone marrow (BM)(n = 7), cord blood (CB)(n = 11) and G-CSF-mobilised peripheral blood samples (mPBP, patient (n = 11); or mPBD, donor (n = 5)). (*** p < 0.001, Mann-Whitney test).