Immunophenotypical and functional heterogeneity of dendritic cells generated from murine bone marrow cultured with different cytokine combinations: implications for anti-tumoral cell therapy

Abstract

Dendritic cells (DC) are professional antigen-presenting cells that can be used as immune adjuvant for anti-tumoural therapies. This approach requires the generation of large quantities of DC that are fully characterized on the immunophenotypical and functional levels. In a murine model, we analysed the in vitro effects of granulocyte-macrophage colony-stimulating factor (GM-CSF) alone or combined with interleukin-4 (IL-4) or Flt3 ligand (Flt3-L) on the number, immunophenotype and functions of bone marrow-derived DC. In GM-CSF cultures, we have identified two populations based on their level of expression of major histocompatibility complex (MHC) class II molecules: MHC-IIhi cells, exhibiting the typical morphology and immunophenotype of myeloid DC (CD11c+ 33D1+ DEC-205+ F4/80+), and MHC-IIlo cells, heterogeneous for DC markers (30% CD11c+; 50% 33D1+; DEC-205-; F4/80+). The addition of Flt3-L to GM-CSF induced a twofold increase in MHC-IIhi DC number; besides, the MHC-IIlo cells lost all DC markers. In contrast, after addition of IL-4 to GM-CSF, the two populations displayed a very similar phenotype (CD11c+ 33D1- DEC-205+ F4/80-), differing only in their expression levels of MHC class II and costimulatory molecules, and showed similar stimulatory activity in mixed leucocyte reaction. We next analysed the migration of these cultured cells after fluorescent labelling. Twenty-four hours after injection into the footpads of mice, fluorescent cells were detected in the draining popliteal lymph nodes, with an enhanced migration when cells were cultured with GM-CSF+Flt3-L. Finally, we showed that MHC-IIhi were more efficient than MHC-IIlo cells in an anti-tumoral vaccination protocol. Altogether, our data highlight the importance of characterizing in vitro-generated DC before use in immunotherapy.

Figure 1

FACS analysis of bone marrow cells cultured in GM-CSF. Bone marrow cells cultured for 6 days in the presence of GM-CSF were stained with mAb for MHC class II molecules for flow cytometry analysis. (a) Histogram showing two populations expressing MHC class II molecules at high (MHC-II^hi) and low levels (MHC-II^lo). (b) and (c) Histograms representing the size (FSC) and the structural features (SSC) of gated MHC-II^hi (thick line) and MHC-II^lo (thin line) cells. Few MHC class II-negative cells were observed in this experiment.

Figure 4

Cell surface immunophenotype of MHC class II-expressing cells obtained after 6 (a) and 9 days (b) of culture in GM-CSF alone (upper row) or in combination with Flt3-L (middle row) or IL-4(lower row). At day 6 of culture, cells were sorted by flow cytometry according to their high or low expression of MHC class II molecules. Immediately and after 3 days of culture (day 9), cells were double-stained for MHC class II molecules and for additional indicated markers. Histograms: dotted line, negative controls labelled with the same amount of isotype-matched control mAb; thick line, MHC-II^hi cells; thin line, MHC-II^lo cells. Representative data from three independent experiments performed on unsorted cells are shown.

Figure 2

Morphological features of MHC-II^hi(a) and MHC-II^lo (b) cell populations after 9 days of culture. MHC-II^hi cells have long and numerous cellular processes and spontaneously form aggregates in culture; in MHC-II^lo cells the processes are short or absent. Magnification ×200.

Figure 3

Cell numbers in culture with GM-CSF, GM-CSF+Flt3-L, GM-CSF+IL-4 after 6 (a) and 9 (b) days. The percentages of MHC-II^hi and MHC-II^lo cells were determined by flow cytometry analysis after double-staining for MHC class II molecules and CD11c markers. These percentages allowed calculation of the number of cells obtained per well for each population. The number of cells of each population is expressed/two legs (tibias and femurs). Mean±SD of six and five independent experiments at day 6 and day 9, respectively. *Significantly different compared to GM-CSF alone.

Figure 5

MLR with sorted MHC-II^hi(•) and MHC-II^lo (○) cells after a 6 (upper panel) or 9-day (lower panel) culture in GM-CSF alone (left panel) or in combination with Flt3-L (middle panel) or IL-4 (right panel). Cells from DBA/2 mice were sorted at day 6 and subcultured until day 9 as described in Fig. 2. Graded numbers of cells were irradiated and cocultured with 10⁵ responding LN cells from C57BL/6 mice. Proliferative responses were measured (c.p.m.) on day 4, which corresponds to the peak of response under our experimental conditions.

Figure 6

FACS analysis of popliteal LN cells 24 hr after injection of cells cultured in GM-CSF, GM-CSF+Flt3-L, or GM-CSF+IL-4 for 6(upper panel) or 9 days (lower panel). Cells were stained with PKH2 and injected s.c into the footpad of syngeneic mice. Twenty-four hours later, the popliteal draining LN were harvested and digested by collagenase treatment. LN cells (1·5×10⁵) were analysed by flow cytometry for their size (forward scatter) and PKH2 green fluorescence. Controls were popliteal LN cells from non-injected mice. Representative data from one of three independent experiments performed with three or four mice per group.

Figure 7

Effect of vaccination with MHC-II^hi and MHC-II^lo cell populations generated in GM-CSF culture pulsed with hCD4 soluble antigen. Survival of mice vaccinated with sorted hCD4-pulsed MHC-II^hi and MHC-II^lo cell populations, unsorted, pulsed or unpulsed BM-derived cells, after challenge with L1210/hCD4 tumour cells (five mice/group).