LentiBOOST® in combination with Polybrene enables transduction of murine BMDCs with HAdV5.
To overcome the low transduction efficacy due to the lack of CAR expression on target cells, or the species-specificity of the HAdV serotype, we examined the transduction agent LentiBOOST® combined with Polybrene in the context of HAdV transduction, of human and murine primary cells. Therefore, we first transduced murine immature and mature BMDCs with a HAdV5 coding for GFP at MOIs from 40 to 500, either in the presence or absence of LentiBOOST® /Polybrene (LeB/ PB; Fig. 1). As a negative control, cells were either mock-transduced or transduced with a HAdV5, encoding for the renilla firefly luciferase (HAdV5Luc1). As assessed by flow cytometry 48 hours afterwards, percentage of GFP+ iBMDCs increased from 2–9% (MOI40), 3–27% (MOI100), 5–42% (MOI200) and 11–54% (MOI500), using LeB/ PB (Fig. 1a). No significant cytotoxicity was observed, except for the highest MOI used in combination with LeB/ PB (Fig. 1b). Similar results were obtained using mBMDCs (Fig. 1c, d). Here, the percentage of GFP+ cells was enhanced by the combination LeB/ PB with increasing MOIs, from 3–15% (MOI40), and from 24–54% (MOI500), while survival of mBMDCs was unaltered (Fig. 1d). Moreover, LeB/ PB did not affect the morphology of either iBMDCs or mBMDCs (Fig. 1e, f).
In summary, LentiBOOST® combined with Polybrene breaks species-specific HAdV5 infection barriers of murine immature and mature BMDCs, without toxic side effects. As MOIs 200 and 500 of HAdV5GFP were shown to be most efficient to transduce immature and mature BMDCs, these were used for the following experiments.
LentiBOOST®/ Polybrene does not alter the phenotype and function of murine BMDCs.
As DCs rapidly react on environmental changes, we next investigated the influence of the LentiBOOST®/ Polybrene regarding the phenotype and function of immature and mature BMDCs, 48 hours after transduction with HAdV5Luc1 (MOI500) and HAdV5GFP (MOI200 and MOI500), in comparison to cells inoculated with adenovirus alone. Flow cytometric analyses revealed no differences regarding the expression of classical DC cell surface markers, such as CD25, CD80, CD83, CD86 and MHC class II, on iBMDCs (Fig. 2a), compared to cells transduced in the absence of LentiBOOST®/ Polybrene (-LeB/ PB). Interestingly, the usage of a GFP encoding adenovirus induced a strong maturation of iBMDCs, independent of the used MOI, which was not observed for firefly luciferase encoding HAdV5 (HAdV5Luc1) transduced cells, or mock-treated cells (Fig. 2a). Expression of CD25, CD80 and CD86, on the cell surface of mature BMDCs, was not affected by LentiBOOST®/ Polybrene (Fig. 2b). Induction of BMDCs maturation was accompanied by the secretion of pro-inflammatory cytokines, like MCP-1, TNF-α, IL-6 and IFN-β, while mock- or HAdV5Luc1 treated iBMDCS produced only very low -, or almost no cytokines (see Supplementary Fig. S1a online). Additionally, HAdV5GFP + LeB/ PB led to an increased production of IL-6 and IFN-β by iBMDCs, when compared with HAdV5GFP – LeB/ PB (see Supplementary Fig. S1a online). A slightly altered secretion of MCP-1, IL-1α and IL-6 was also observed for mBMDCs (see Supplementary Fig. S1b online).
To address the functionality of these differentially treated mBMDCs, mixed lymphocyte reactions (MLR) were performed (Fig. 3a, b). Here, LentiBOOST®/ Polybrene (“Mock” + LeB/ PB) or in combination with HAdV5Luc1 (MOI 200 and MOI 500), did not affect the capability of mBMDCs to prime allogeneic T cells. By contrast, transduction of mBMDCs with HAdV5GFP in the presence of LentiBOOST®/ Polybrene led to a stronger proliferation of T cells at a DC:T cell ratio of 1:10 (1.3 fold MOI200/ 1.7 fold MOI500), and 1:33 (1.5 fold MOI200/ 1.7 fold MOI500). This effect was not observed for HAdV5GFP transduced BMDCs missing the LeB/ PB. Next, we examined the portfolio of classical T cell derived cytokines in the supernatants of the co-cultures. We observed no differences between BMDCs transduced in the absence (-LeB/ PB) or presence (+ LeB/ PB), to induce classical Th1/ Th2 associated cytokines IL-2 and IL-6 (Fig. 3c), Th1 cytokines IFN-γ and TNF-α (Fig. 3d), Th2 cytokines IL-4 and IL-13 (Fig. 3e), or Th17 cytokines IL-17A and IL-22 (Fig. 3f).
Taken together, phenotype and function of BMDCs are not altered by LentiBOOST®/ Polybrene, although a slight increase (i) in the secretion of pro-inflammatory cytokines by iBMDCs, and (ii) the capability of mBMDCs to prime T cells was observed, partially depending on the transgene encoded by the adenovirus.
LentiBOOST®/ Polybrene increases transduction efficiency of human monocyte-derived DCs at low HAdV5 virus titres.
Although lacking CAR expression, human DCs can be transduced with HAdV5 at high titres > MOI500. Since high virus titres are accompanied by toxicity, followed by loss of function, we next assessed if LentiBOOST®/ Polybrene can improve HAdV5-mediated gene delivery to human DCs. Flow cytometric analyses, 48 hours after transduction of immature (i)DCs and mature (m)DCs with HAdV5GFP at different MOIs, revealed that a MOI of 10 is already sufficient to infect ~ 60% of iDCs, while without LentiBOOST®/ Polybrene only 1% of cells were GFP-positive (Fig. 4a). Using a MOI of 100 or 200, > 80% of cells were GFP-positive, while only 22% or 42% of cells were transduced in the absence of LentiBOOST®/ Polybrene, respectively. Toxicity was moderate, with a decrease of living cells from 96–80% (MOI100), or 92–65% (MOI200; Fig. 4b). Transduction efficacy of mDCs increased from 1–20% (MOI 10), 1–49% (MOI 20), 13–67% (MOI40), 43–84% (MOI100) and from 63–84% (MOI200), using the LeB/ PB (Fig. 4c). Again, survival of mDCs was not impaired (Fig. 4d). Moreover, iDCs and mDCs showed a similar morphology and cell clustering may be increased in the presence of LentiBOOST®/ Polybrene, independent of the adenovirus (Fig. 4e and f). Overall, LentiBOOST® combined with Polybrene efficiently transduces iDCs and mDCs, using a MOI of 10 or 40, respectively, with an optimal MOI of 100.
LentiBOOST®/ Polybrene induces maturation of immature human DCs but does not alter DC function.
Next we analysed, if LentiBOOST®/ Polybrene and the observed enhanced infection rate influence the phenotype and function of human DCs. Thus, iDCs and mDCs, either mock-treated, or transduced with HAdV5Luc1 (MOI200), or HAdV5GFP (MOI 100 and 200), were analysed by flow cytometry, 48 hours post infection. In contrast to murine iBMDCs, human iDCs exposed to LentiBOOST®/ Polybrene showed a significant upregulation of maturation markers CD83 and CD86, while CD25, CD80 and HLA-DR were not statistically altered (Fig. 5a). For HAdV5GFP the median was even higher when compared to mock- or HAdV5Luc1 controls. Interestingly, we observed almost no differences regarding the cytokine production, except for IL-8, which was clearly upregulated in the presence of the LentiBOOST®/ Polybrene (see Supplementary Fig. S2a online). However, inoculation of mDCs with LentiBOOST®/ Polybrene did not alter their expression profiles in comparison to control mDCs (Fig. 5b), nor changed their cytokine production (see Supplementary Fig. S2b online). Moreover, median fluorescence intensity was strongly increased for all analysed cell surface markers, including CD80, CD83, CD86 and HLA-DR, when compared with LentiBOOST®/ Polybrene treated iDCs.
Next, to analyse the mDCs-mediated allogeneic T cell proliferation, MLR assays were performed. Results revealed that neither LentiBOOST®/ Polybrene nor the amount of adenovirus used influenced the T cell stimulatory capacity of mDCs (Fig. 6a, b). When analysing the supernatants derived from the DC - T cell co-cultures for their content of classical T-cell derived cytokines, like IL-2, IL-6, IFN-γ, IL-5, IL-13 and IL-22, no differences between + LeB/ PB and –LeB/ PB were observed (Fig. 6c-f). However, co-cultures of T cells with HAdV5GFP transduced mDCs showed reduced levels of IL-5, IL-6, IL-13 and IL-22, independent of the LentiBOOST®/ Polybrene.
Taken together, HAdV5-mediated gene transfer, in the presence of LentiBOOST®/ Polybrene, does not significantly interfere with the DC phenotype and their function.
Efficient HAdV5-mediated gene transfer into monocytes by LentiBOOST®/ Polybrene allows for proper subsequent iDCs and mDCs differentiation.
Similar to murine BMDCs, human monocytes are poorly permissive to HAdV5, even at high titres. Hence, we transduced freshly isolated human monocytes (see Supplementary Fig. S3a online) with HAdV5GFP at a MOI of 100 or 200, either in the absence or presence of LentiBOOST®/ Polybrene. Twenty-four hours afterwards, cells were analysed by flow cytometry. As depicted in Fig. 7a, the transduction efficacy increased from 7–36% (MOI100) and 15–36% (MOI200), when applying LentiBOOST®/ Polybrene to the cell culture. Notably, mean GFP expression was upregulated from 227 to 3591 (MOI100) and 588 to 4264 (MOI200), while no toxicity (Fig. 7a), no changes in the composition of monocyte subpopulations (see Supplementary Fig. S3b online), or cytokine secretion (see Supplementary Fig. S4a online), were observed. Next, transduced monocytes were differentiated into iDCs and mDCs and analysed regarding their phenotype and function. Flow cytometric analyses of day four iDCs, differentiated from transduced monocytes, using IL-4 and GM-CSF, revealed no loss of GFP positivity, in comparison to monocytes, although median values were diminished (Fig. 7b). Again, survival of cells was shown to be > 92%. Further exploration of these cells displayed a typical immature DC-phenotype regarding the expression of CD25, CD80, CD83, CD86 and HLA-DR, in comparison to the “Mock”-control, independent of the presence or absence of LentiBOOST®/ Polybrene (Fig. 7c). In addition, cytokine profiles were unaltered (see Supplementary Fig. S4b online).
Next, mDCs were generated from these iDCs by adding a maturation cocktail, consisting of IL-1β, IL-6, PGE2 and TNF-α for another 24 hours. While GFP expression and survival of those cells was comparable to iDCs shown before (Fig. 7d), matured DCs highly upregulated typical DC maturation markers, such as CD25, CD80, CD83, CD86 and HLA-DR (Fig. 7e). Importantly, the LentiBOOST®/ Polybrene did not affect this maturation process. However, exposure of monocytes to HAdV5GFP alone led to a 2.6- (MOI100) or 2.4 (MOI200) fold increased expression of CD25 on mDCs, which was not observed when LeB/ PB was present during transduction. Moreover, transduction with HAdV5GFP affected upregulation of CD83 in a virus concentration dependent manner, which in contrast to CD25 was independent of the usage of LentiBOOST®/ Polybrene. Interestingly, in the supernatants of mock-treated cells, without LeB/ PB, we found strongly increased IL-6, TNF-α and MCP-1 concentrations (see Supplementary Fig. S4c online).
Finally, we investigated the function of these mDCs, using an allogeneic MLR assay and in addition we used them to prime autologous T cells, in a tumour-antigen-specific manner (Fig. 8). Regarding MLR assays, mock- and HAdV5GFP-treated monocytes -/+ LeB/ PB, differentiated into mDCs were co-cultured with allogeneic T cells at different ratios, before T cell proliferation was assessed by thymidine incorporation (Fig. 8a). Although we observed a HAdV5GFP- and virus concentration- mediated effect regarding the capacity of mDCs to prime naïve T cells, no LentiBOOST®/ Polybrene dependent alterations were observed. In addition, transduced cells (-/+ LeB/ PB), were still able to induce a high and efficient T-cell proliferation accompanied by an unaltered IL-2 secretion (Fig. 8a, b).
To prime autologous T cells, mDCs were additionally electroporated with RNA coding for the tumour antigen MelanA (MelA) or loaded with a MelA specific peptide. Survival of DCs four hours after electroporation varied from 40–60% (Fig. 8c). Next, MelA peptide loaded, MelA RNA or control electroporated (“no RNA”) DCs, were used to stimulate bulk autologous CD8+ T cells for one week, before the fraction of MelA-specific T cells was determined by tetramer-staining. As shown in Fig. 8d and S5 (see Supplementary Information online), MelA electroporated and MelA-peptide loaded DCs induced antigen-specific CD8+ T cells, compared to control DCs (“no RNA”), even when they had been transduced with HAdV5 and treated with LentiBOOST®/ Polybrene before. Noteworthy, the numbers of MelA+ CD8+ T cells vary in a donor-dependent manner (see Supplementary Fig. S5 online), which is rather common when cells derived from different healthy donors are used. However, these data clearly indicate that LentiBOOST®/ Polybrene does not impair the capability of DCs to express, process, and present a tumour specific antigen in an MHC-class I-restricted manner.
In summary, using LentiBOOST®/ Polybrene not only allows for the highly efficient transduction of human monocytes with HAdV5, but additionally for the differentiation into iDCs and functional mDCs.