Dendritic cells (DCs) play critical roles in the induction and regulation of the innate and adaptive immune responses. Human blood DCs can be classified into plasmacytoid dendritic cell (pDC) and myeloid dendritic cell (mDC). In general, pDC is defined as lineage (Lin)-HLA-DR (DR)+CD123+CD11c-, and mDC is defined as Lin-DR+CD123+CD11c+. PDCs are a specific type of dendritic cells that is found in an immature form in the peripheral blood and that is the major interferon-alpha producing cell in response to viruses. Blastic plasmacytoid dendritic cell neoplasm (BPDCN) is a rare hematological malignancy that has a putative plasmacytoid dendritic cells origin. Unlike blood pDCs, the specific feature of BPDCN is the positive expression of CD56. In addition to these markers, BPDCNs can express various antigens, such as CD2, CD10, CD13, CD33 and even CD11c, that cause immunophenotypical diversity among cases. The goal of this study was to clarify the normal counterpart of BPDCN by analyzing the characteristics of CD56-positive blood Dendritic-like Cells (DLCs).
Human peripheral blood mononuclear cells (PBMNCs) were isolated by gradient centrifugation from healthy volunteers, and CD3, CD14, CD16 and CD19 antibodies were used as a lineage cocktail. We defined CD56+pDC-like cells (pDLCs) as Lin-DR+CD56+CD123+ cells, CD56+mDC-like Cells (mDLCs) as Lin-DR+CD56+CD123-CD11c+ cells, pDCs as Lin-DR+CD56-CD123+CD11c-cells and mDCs as Lin-DR+CD56-CD123+CD11c+cells. In some experiments, cells were purified from PBMNCs using a cell sorter. Sorted cells were analyzed for mRNA levels of toll-like receptors (TLRs), cytokines and transcriptional factors. Phagocytic activity and mixed lymphocyte reactions were analyzed by flow cytometry. Sorted cells were also analyzed after 4–6 days of culture with Fms-like tyrosine kinase 3 ligands (Flt3-L) and granulocyte macrophage colony-stimulating Factor (GM-CSF).
PBMCs comprised a small population of each cell type: 0.03% of CD56+pDLCs, 0.35% of CD56+mDLC, 0.93% of pDC 0.93%, and 0.60% of mDC. CD56+pDLCs had oval or U-shaped nuclei with condensed chromatin, and perinuclear halo, which is feature of pDC, was clearly observed in the cytoplasm. CD11c expression in CD56+pDLCs was lower than that in mDCs but higher than that in pDCs. CD56+pDLCs were not Natural Killer (NK) cells, as there was no expression of CD122 or other NK-specific antigens. Meanwhile, CD56+pDLCs had clear expression of BDCA2 and BDCA4, suggesting that this population was closely related to pDCs. Real-time quantitative (RQ) PCR assay revealed that TLRs were expressed in an intermediate level between pDCs and mDCs in CD56+pDLCs (CD56+pDLC vs. pDC vs. mDC: TLR2, 0.17 vs. 0.09 vs. 1.13; TLR4, 0.14 vs. 0.06 vs. 0.53; TLR7, 0.67 vs. 16.70 vs. 0.30; TLR9, 3.73 vs. 72.41 vs. 0.18). Expression of the transcription factors, E2-2, Irf8 and SpiB, in pDCs was higher than that in CD56+pDLCs, but lower than that in mDCs (CD56+pDLC vs. pDC vs. mDC: E2-2, 16.78 vs. 118.69 vs. 1.45; Irf8, 1.73 vs. 9.07 vs. 0.55; SpiB, 0.14 vs. 0.52 vs. 0.02). RQ−PCR after CpG stimulation revealed that CD56+pDLCs had lower interferon–alpha production when compared with pDCs (5.7405 vs. 360.881). Phagocytic capacity of CD56+pDLCs was lower than that of mDC or pDC (1.96% vs. 4.32 % vs. 52.6% for FITC-dextran positive cells in CD56+pDLCs vs. pDCs vs. mDCs). Allogeneic T cells proliferated less efficiently after culture with CD56+pDLCs than they did after culture with pDC. After in vitro culture with Flt3L and GM-CSF, the percentage of BDCA1-positive cells increased from 2.75% to 62.9%.
CD56+pDLCs were rare population in PBMNCs. Their phenotype and function were similar to pDCs, in part, but they expressed myeloid antigens and had lower function of phagocytosis and cytokine production than pDCs. In vitro culture suggested plasticity in the immunophenotype of CD56+pDLCs when compared with pDC and mDC. Collectively, these data suggest that CD56+pDLCs is a distinct new population of DCs that possesses a high degree of plasticity. These immunophenotypic characteristics and plasticity may influence the immunophenotypic diversity of BPDCNs.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.