Dendritic Cell Subsets in Bone Marrow and Peripheral Blood of Patients with Myelodysplastic Syndromes Display Numeric and Functional Defects

Introduction

Immune response plays an important role in the pathogenesis and progression of myelodysplastic syndromes (MDS). It has been shown that a consistent immunological signature in MDS comprises an increase in Th17 cells in low risk MDS and an expansion of Tregs in high risk subtypes. The presence of high Treg numbers in lower risk stages is an independent prognostic value and is associated with poor survival. Whereas the immune response in low risk MDS is pro-inflammatory, high risk MDS is characterized by immunosuppression which could contribute into the expansion of the dysplastic clone and malignant transformation. The exact role of dendritic cells (DC) in inducing this immune dysfunction is yet to be understood. The aim of this comparative study was therefore to investigate the frequency and function of different DC subsets in the bone marrow (BM) and peripheral blood (PB) of MDS patients compared to healthy donors (HD).

Patients and methods

The enumeration of DC subsets was performed in 110 MDS and in 19 HD bone marrow samples. In peripheral blood 34 MDS and 18 HD samples were investigated. DC were identified by the expression of HLA-DR and the lack of the lineage markers CD14 and CD19. They were further subdivided into CD303+ plasmacytoid DC (pDC), and the CD11c+ myeloid subsets CD1c+ myeloid DC 1 (mDC1), CD141hi myeloid DC 2 (mDC2) and CD16+ 6-Sulfo LacNac (Slan) DC. Furthermore, their ability to upregulate co-stimulatory molecules was assessed by flow cytometry and the secretion of cytokines in response to TLR ligands was analyzed using Cytometric Bead Array (CBA). Their capacity to induce allogeneic T cell proliferation was evaluated in a mixed leukocyte reaction (MLR).

Results

The frequencies of pDC, mDC1 and mDC2 but not of SlanDC were significantly lower in patients' BM compared to HDs' BM (pDC 0.33% vs 0.72%, p=0.01; mDC1 0.30% vs 0.73%, p<0.001; mDC2 0.01% vs 0.05%, p<0.0001; SlanDC 0.10% vs 0.13%, p=0.20). In PB all DC subsets were significantly lower in patients' samples compared to HDs' samples (pDC 0.15% vs 0.28%, p=0.02; mDC1 0.12% vs 0.61%, p<0.0001; mDC2 0.003% vs 0.045%, p<0.0001; SlanDC 0.19% vs 0.46%, p=0.04). A positive correlation was found between the frequencies of DC subsets in a paired assessment of PB and BM of 30 MDS patients (pDC r=0.77, mDC1 r=0.57, mDC2 r=0.34, SlanDC r=0.60). Of note, in a more advanced disease state DC subsets showed a progressive decline in frequency. To investigate DC function they were isolated or FACS sorted from HD BM (n=3) or patient BM (n=4). Based on their TLR expression profile a combination of LPS and R848 has been used to stimulate DC overnight. In HD upregulation of the co-stimulatory molecules CD80 and CD86 was observed after this stimulation. However, MDS DC stimulated with the same combination of TLR ligands showed a reduced upregulation of CD80 and CD86 (median fluorescence intensity (MFI) CD80: SlanDC 2323 vs 6254, MFI CD86: mDC1 2785 vs 4608; SlanDC 2344 vs 5087). They also produced lower levels of pro-inflammatory cytokines compared to HD DC as measured in culture supernatants (mDC1 TNF [pg/ml] 997 vs 2980, IL-6 [pg/ml] 843 vs 2806, IL-8 [pg/ml] 966 vs 5549 and IL-12p70 [pg/ml] 105 vs 1115). Additionally, after 5 days co-culture of CFSE labeled T cells and DC, mDC1 and SlanDC derived from MDS BM displayed a defective induction of allogeneic CD4+ and CD8+ T cell proliferation compared to HD (mDC1 3.08% vs 10.49% CD4+ T cell proliferation and 4.54% vs 15.07% CD8+ T cell proliferation; SlanDC 1.28% vs 5.23% CD4+ T cell proliferation and 1.50% vs 8.67% CD8+ T cell proliferation).

Conclusion

Our data clearly point to both a numeric and functional impairment of myeloid DC subsets in BM as well as in PB of patients with MDS. A further decline of DC frequencies in high risk MDS may support the therapeutic targeting of DC in the BM microenvironment of these patients to redress immune dysfunction and possibly prevent further progression to acute myeloid leukemia. Extensive research is needed to further delineate the differences in DC function and immune composition in MDS risk categories.