Highly efficient transduction of human plasmacytoid dendritic cells without phenotypic and functional maturation

Abstract

Background: Gene modified dendritic cells (DC) are able to modulate DC functions and induce therapeutic immunity or tolerance in an antigen-specific manner. Among the different DC subsets, plasmacytoid DC (pDC) are well known for their ability to recognize and respond to a variety of viruses by secreting high levels of type I interferon.

Methods: We analyzed here, the transduction efficiency of a pDC cell line, GEN2.2, and of pDC derived from CD34+ progenitors, using lentiviral vectors (LV) pseudotyped with different envelope glycoproteins such as the vesicular stomatitis virus envelope (VSVG), the gibbon ape leukaemia virus envelope (GaLV) or the feline endogenous virus envelope (RD114). At the same time, we evaluated transgene expression (E-GFP reporter gene) under the control of different promoters.

Results: We found that efficient gene transfer into pDC can be achieved with VSVG-pseudotyped lentiviral vectors (LV) under the control of phoshoglycerate kinase (PGK) and elongation factor-1 (EF1alpha) promoters (28% to 90% of E-GFP+ cells, respectively) in the absence of phenotypic and functional maturation. Surprisingly, promoters (desmin or synthetic C5-12) described as muscle-specific and which drive gene expression in single strand AAV vectors in gene therapy protocols were very highly active in pDC using VSVG-LV.

Conclusion: Taken together, our results indicate that LV vectors can serve to design pDC-based vaccines in humans, and they are also useful in vitro to evaluate the immunogenicity of the vector preparations, and the specificity and safety of given promoters used in gene therapy protocols.

Figure 1

Transduction efficiencies of GEN2.2. The pDC cell line, GEN2.2, was non-transduced (NT) or transduced with E-GFP encoding vectors then analysed 5 days posttransduction. GEN2.2 were gated in forward/side scatter, then analyzed for the expression E-GFP by flow cytometry. (A) GEN2.2 were transduced by LV with a PGK promoter pseudotyped with either VSVG and RD114 envelopes at a MOI of 18 or with the GaLV envelope at a MOI of 9. (B) GEN2.2 were transduced with VSVG pseudotyped-LV with a PGK, EF1, desmin or C5–12 promoter, at a MOI of 18. (C) GEN2.2 were transduced by rAAV of serotype 1, 2 or 5 with a CMV promoter, with the number of viral genomes/cell indicated. Results are expressed as mean percentage of cell +/- SD over the number of independent experiments indicated.

Figure 2

Transduction efficiencies of CD34-pDC. CD34-pDC were non-transduced (NT) or transduced with E-GFP encoding vectors at day 6 after the induction of differentiation, then cultured for 6 additional days. pDC were gated in forward/side scatter, then analyzed for the expression of E-GFP by flow cytometry. (A) CD34⁺ progenitors were transduced by LV with a PGK promoter pseudotyped with either VSVG and RD114 envelopes at a MOI of 18 or with the GaLV envelope at a MOI of 9. (B) CD34⁺ progenitors were transduced with VSVG pseudotyped-LV with a PGK, EF1, desmin or C5–12 promoter, at a MOI of 18. (C) CD34⁺ progenitors were transduced by rAAV of serotype 1, 2 or 5 with a CMV promoter, with the number of viral genomes/cell indicated. Results are expressed as mean percentage of cell +/- SD over the number of independent experiments indicated.

Figure 3

Immunophenotype of transduced pDC. Comparative phenotypes of transduced and untransduced GEN2.2 in absence of maturation agent, at day 5. Overlay histograms show the expression of CD123 or HLA-DR for untransduced (thin line), total transduced (thick line) and E-GFP⁺ gated (green line) GEN2.2, versus isotype-matched controls (dotted line). (A) GEN2.2 transduced by LV with a PGK promoter pseudotyped with either VSVG, RD114 or GaLV envelopes. (B) GEN2.2 were transduced with VSVG pseudotyped-LV with a PGK, EF1, desmin or C5–12 promoter (C) GEN2.2 were transduced by rAAV of serotype 1, 2 or 5 with a CMV promoter. The results are representative of at least 4 experiments.

Figure 4

Transduction of GEN2.2 does not induce maturation. Comparative phenotype of transduced and non transduced GEN2.2. Overlay histograms show the expression of relevant antigens for untransduced (thin line) and transduced (thick line) with LV-VSVG/PGK or rAAV2/2, versus isotype-matched controls (dotted line). The results are representative of at least 3 experiments.

Figure 5

CpG induced maturation of transduced pDC. Comparative phenotype of transduced GEN2.2, 6 days posttransduction, in the absence and presence of CpG for 24 hours. Overlay histograms show the expression of relevant antigens for transduced GEN2.2 cultured without CpG (thin line), with CpG (thick line) and with CpG and gated on E-GFP⁺ (green line) versus isotype-matched controls (dotted line). (A, B, C) Transduced GEN2.2 vectors are the same ones as those described in figure 2. Values indicated are MFI of the transduced populations cultured without CpG versus transduced populations cultured with CpG. The results are representative of at least 4 experiments.

Figure 6

T-cell stimulatory capacity of non-transduced and transduced GEN2.2 in mixed lymphocyte alloreactions. Day 5 transduced GEN2.2 were matured in CpG for 24 hours, before cell sorting on an E-GFP expression basis. (A-E) Total non-transduced (NT), E-GFP^- and E-GFP⁺cell sorted GEN2.2 transduced by the same LV as those described in figure 3 were incubated with allogeneic T cells stained with CFSE. (F) Total non-transduced (NT) and rAAV2/1, rAAV2/2 or rAAV2/5 transduced unsorted GEN2.2 were incubated with allogeneic T-cells stained with CFSE. After 4 days of co-culture, percentages of CD3⁺ dividing T cells measured by flow cytometry were linearly correlated with the loss of CFSE fluorescence. Dot plots inserted in graphs show one representative CFSE profile at the ratio 3/1 for GFP⁺ cells. The data are shown as the means of 3 independent experiments.

Figure 7

IFN-α production by pDC. GEN2.2 and day 6 CD34-pDC were non transduced (NT) or transduced by LV-VSVG at an MOI of 18 (LV-VSVG), then 6 days later, the IFNα production was measured in cell culture supernatants before or after maturation in CpG, for 24 hours. The data are shown as the means of 3 independent experiments.

Figure 8

CD8⁺ T cell clone activation by LV transduced pDC. In vitro antigen presentation capacities of LV transduced HLA-A2 pDC cells and Mo-DC. Cells were transduced with LV encoding the MART-1 peptide under the control of the PGK promoter. (A) Mature non-transduced (NT) and transduced (VSVG-PGK-MART-1) GEN2.2 or (B) CD34-pDC and Mo-DC were co-cultured with the HLA-A2 restricted CD8⁺ T-cell clone specific for the MART-1 peptide (LT12) stained with CFSE. After 5 days of co-culture, percentages of CD8⁺ dividing T-cells measured by flow cytometry were linearly correlated with the loss of CFSE fluorescence. The data in panel A are shown as the mean of triplicate and represent one out of 3 independent experiments whereas the data in panel B were performed once.