High-Level Antigen Expression and Sustained Antigen Presentation in Dendritic Cells Nucleofected with Wild-Type Viral mRNA but Not DNA

Plasmid and mRNA generation.Generation of the pSP73/GFP/A64 and pSP73/SIVmac239Gag/A64 plasmids and in vitro transcription of mRNA were performed as described previously (33). pSP73/SIVmac239Gag/A64 encodes a wild-type, non-codon-optimized Gag sequence isolated from a rhesus macaque 2 weeks postinfection with the SIVmac239 SIV isolate (33). pSP73/SIVmac239Gag/A64 was used as a template for the amplification of Gag by PCR using the following forward and reverse primers, respectively: pGagN1F, 5′GCGCTCGAGGCCACCATGGGCGTGAG3′; and pGagN1R, 5′CGCGCGGCCGCTTACTTGCCCAACTGCATGTAG 3′. pEGFP-N1 DNA was digested with XhoI and NotI, and the larger vector band was retrieved by gel purification. XhoI- and NotI-digested gag PCR product was inserted to the XhoI- and NotI-digested pEGFP-N1 DNA to generate pSIVmac239Gag-N1 encoding the wild-type sequence.

Cells.DC were cultured from the purified blood monocytes of SIV-naïve rhesus macaques or healthy human volunteers as described previously (6). In some experiments, DC were matured for 24 h or 48 h with 3 μg/ml recombinant trimeric CD40L (Immunex, Seattle, WA) as described previously (6). Approval was obtained from the institutional review board prior to experiments involving human samples and from the institutional animal use and care committee for all experiments involving rhesus macaque samples. K562 cells were grown in Iscove's modified Dulbecco's medium (HyClone, Logan, UT) supplemented with 10% fetal bovine serum.

Transfection of K562 cells and DC.K562 cells were washed twice with Opti-MEM (Gibco Invitrogen Corporation, Frederick, MD) and transfected with TransFast transfection reagent (Promega, Madison, WI) or TransMessenger reagent (Qiagen, Valencia, CA) by adding 2 or 4 μg GFP mRNA or DNA in Opti-MEM with other reagents provided by the manufacturers to 1 × 10⁶ cells as described previously (33). For electroporation, 1 × 10⁶ K562 cells were electroporated with 10 μg mRNA or DNA in a total volume of 250 μl in a 0.4-cm cuvette using the Gene Pulser II electroporation system (Bio-Rad, Hercules, CA) with voltage/capacitance settings of 260 V/150 μF or 300 V/150 μF, respectively (50). Immediately after electroporation, cells were incubated for 5 min on ice. For nucleofection, 10 μg of DNA or mRNA was added to 1 × 10⁶ K562 cells in Opti-MEM in a final volume of 100 μl and transfected in a 2-mm-wide electroporation cuvette (BTX, San Diego, CA) using the T-16 program of the amaxa nucleofector (Amaxa, Koln, Germany). Smaller amounts of mRNA and DNA were used for lipofection than for electroporation and nucleofection as toxicity was noted at higher amounts with the former methods (data not shown). K562 cells were cultured in Iscove's modified Dulbecco's medium for 24 h at 37°C following transfection. DC were transfected as immature cells (day 5 of culture) or as mature cells after a 24-h incubation with CD40L (day 6 of culture). Lipofection using TransFast or TransMessenger reagents was done as for K562 cells. For electroporation, 1 × 10⁶ to 2 × 10⁶ DC in Opti-MEM were mixed with 5 μg to 20 μg mRNA or DNA, depending on the experiment, and electroporated as for K562 cells using voltage/capacitance settings of 300 V/150 μF and 250 V/300 μF, respectively, as described previously (33, 41). Nucleofection of DC was done as described for K562 cells using 5 μg to 20 μg mRNA or DNA and the U-02 or T-01 Amaxa program. Details relating to the electrical parameters of electroporation using the Amaxa nucleofector are proprietary. Nucleofection was performed according to the manufacturer's instructions, with some modifications. In preliminary studies, we determined that nucleofection of cells in Opti-MEM and buffers provided by the manufacturer resulted in similar efficiencies of transfection (data not shown), and we therefore used Opti-MEM for all transfection experiments. Following transfection, DC were cultured in prewarmed complete RPMI medium supplemented with granulocyte-macrophage colony-stimulating factor and interleukin-4 for 24 h or 48 h at 37°C, with and without CD40L, as described previously (33).

Confocal microscopy.DC nucleofected with pEGFP-N1 DNA or gfp mRNA 24 h previously were harvested using 20 mM EDTA and resuspended in phosphate-buffered saline prior to being settled onto glass slides at 37°C for 1 h. Adhered cells were fixed with 2% paraformaldehyde for 15 min and washed in phosphate-buffered saline. Gelvatol was used to apply coverslips to slides. DC were imaged for GFP and differential interference contrast using an Olympus FluoView 500 laser scanning confocal microscope (Olympus, Center Valley, PA). Images were collected using MetaMorph software (Molecular Devices, Sunnyvale, CA).

Flow cytometric analysis.GFP expression in transfected K562 cells and DC and expression of HLA-DR, CD80, CD83, CD86, and CD40 on transfected DC were done as described previously (33).

ELISPOT assay.Gag-specific gamma interferon (IFN-γ) enzyme-linked immunospot (ELISPOT) assays were done as described previously (33). Briefly, immature monkey DC were nucleofected with 10 μg pSIVmac239Gag-N1 DNA or with gag mRNA transcribed from pSP73/SIVmac239Gag/A64 and simultaneously matured with CD40L (3 μg/ml) for 24 h or 48 h prior to incubation with autologous peripheral blood mononuclear cells (PBMC) at a 1:10 ratio. Control cells were nucleofected with pEGFP-N1 DNA or gfp mRNA transcribed from pSP73/GFP/A64 or mock transfected. IFN-γ spot-forming cells were developed and enumerated as described previously (4).

Nucleofection with mRNA is a superior method of introducing transgene into primary human and monkey monocyte-derived DC.We performed a side-by-side comparison of TransFast and TransMessenger lipofection methods along with electroporation and nucleofection using a GFP reporter construct. Cells were transfected with either pEGFP-N1 DNA or in vitro-transcribed gfp mRNA that was generated from the plasmid pSP73/GFP/A64 (33) and monitored for GFP expression at 24 h by flow cytometry. For lipofection methods, we used 4 μg DNA or mRNA as toxicity was noted at higher amounts (not shown), whereas for electroporation and nucleofection we used 10 μg DNA or mRNA. In initial experiments, we used the chronic myelogenous leukemic cell line K562 that is readily transfected with plasmid DNA or mRNA (7, 50). Transfection of K562 cells with EGFP-N1 DNA using the two lipofection methods produced no detectable GFP fluorescence, whereas transfection via electroporation resulted in 16% GFP-expressing cells at 24 h. In contrast, 60% of K562 cells expressed GFP following nucleofection with EGFP-N1 DNA using the T-16 program of the nucleofector device, which was shown in preliminary experiments to generate the highest efficiency of transfection with this cell line (Fig. 1a and data not shown). Transfection with mRNA was significantly more effective than transfection with DNA by all methods, with 8% to 16% GFP expression following lipofection and 86% and 91% of K562 cells expressing GFP following electroporation and nucleofection, respectively (Fig. 1a).

• Open in new tab
• Download powerpoint

FIG. 1.

Evaluation of DNA and mRNA transfection of K562 cells and human and monkey DC. K562 cells (a) or immature human (b) or rhesus macaque (c) monocyte-derived DC were transfected with pEGFP-N1 DNA (top row) (filled histograms) or gfp mRNA (bottom row) (filled histograms) or were mock transfected (empty histograms) using the methods indicated and analyzed by flow cytometry for GFP expression 24 h later. A total of 4 μg DNA or mRNA was used for lipofection with TransFast and TransMessenger reagents, whereas 10 μg DNA or mRNA was used for electroporation and nucleofection. Numbers represent the percentage of cells expressing GFP based on mock transfection.

We next tested the capacities of lipofection, electroporation, and nucleofection to transfect human and monkey immature monocyte-derived DC. In addition to measuring GFP expression at 24 h posttransfection, we assessed the effect of transfection on DC viability by using the standard approach of trypan blue exclusion. Transfection of DC with DNA using lipofection or electroporation produced negligible GFP expression and significant cell death, with viability ranging from 40% to 57% (Fig. 1b and c) (Table 1). Nucleofection with DNA using the U-02 program was more effective at generating GFP-expressing DC than the other methods but was associated with enhanced cell death, with only 24% of DC remaining viable at 24 h posttransfection (Fig. 1b and c) (Table 1), similar to other reports (29). Switching the nucleofector program to T-01 as favored by others (29) did not enhance transfection efficiency, and increasing the quantity of DNA to 20 μg was associated with a significant increase in toxicity (data not shown). Lipofection with mRNA resulted in variable transfection of DC, with maximum expression noted in monkey DC following TransMessenger lipofection, although the proportion of viable cells was only 52% to 55% (Fig. 1b and c) (Table 1). In contrast to K562 cells, electroporation of DC with mRNA was inefficient, with expression levels not exceeding 17% and viability averaging only 46% (Fig. 1b and c) (Table 1). However, nucleofection with mRNA using the U-02 program was a highly effective means of transfecting human and monkey monocyte-derived DC, with the proportion of DC expressing GFP reaching 96% at 24 h posttransfection (Fig. 1b and c) (Table 1). Nucleofection with mRNA maintained the highest cell viability of all methods, although the proportion of viable cells at 24 h posttransfection was still only 68% (Table 1), consistent with results of other reports (27). Similar transfection efficiency and DC viability were noted when nucleofection was performed with a range of mRNA from 5 μg to 20 μg (data not shown).

Flow cytometric analysis of nucleofected cells indicated that GFP expression varied in intensity and uniformity depending on whether DNA or mRNA was used, with DNA producing a wide range of fluorescence and mRNA generating uniform expression in almost all cells (Fig. 1b and c). To evaluate this in more detail, we examined immature monkey DC by confocal microscopy 24 h after DNA and mRNA nucleofection. The majority of mRNA-transfected DC showed a similar intensity of GFP expression when examined individually by microscopy, whereas GFP expression in DNA-transfected DC was highly variable, with some cells having intense fluorescence and a majority having weak or undetectable fluorescence (Fig. 2). Expression of GFP in DNA- or mRNA-transfected DC was cytoplasmic in distribution, as expected (Fig. 2). Similar results were found using human monocyte-derived DC (data not shown). Taken together, these data indicate that nucleofection with mRNA is a superior method of introducing transgenes into DC, both with respect to antigen expression and to cell viability.

• Open in new tab
• Download powerpoint

FIG. 2.

Expression of GFP in mRNA- and DNA-nucleofected DC. Immature monkey monocyte-derived DC were nucleofected with gfp mRNA (a and c) or pEGFP-N1 DNA (b and d) and examined by confocal microscopy 24 h later. Shown are GFP expression (left) and differential interference contrast (right). (a and b) Original magnification, ×200. (c and d) Original magnification, ×1,500.

mRNA nucleofection of DC produces rapid and sustained expression of transgene.A potential limitation of mRNA transfection of DC for immunotherapy is that mRNA is labile in cells and may be degraded rapidly, resulting in limited duration of antigen expression. We therefore assessed the durability of transgene expression in immature monkey DC by harvesting cells at various intervals after DNA or mRNA transfection and determining the proportion of cells expressing GFP by flow cytometry. For these and all other experiments, we focused on nucleofection as the preferred method of transfection. GFP expression following nucleofection of DC with pEGFP-N1 DNA was detectable at 3 h posttransfection and maintained relatively constant levels of expression from 6 h to 48 h posttransfection, after which the proportion of GFP-expressing cells markedly declined (Fig. 3), similar to other reports (29, 45). In contrast, the proportion of monkey DC expressing GFP following nucleofection with in vitro-transcribed gfp mRNA reached near-maximal levels by 3 h and stayed at this high level for 96 h, the duration of the experiment (Fig. 3). Similar rapid and sustained kinetics of transgene expression were noted in other studies following mRNA delivery to human DC using either electroporation or nucleofection (27, 31, 33, 46, 50). These findings indicate that the delivery of mRNA rather than DNA generates a more-durable expression of transgene in monocyte-derived DC.

• Open in new tab
• Download powerpoint

FIG. 3.

High-level and durable GFP expression in monkey DC nucleofected with gfp mRNA but not with pEGFP-N1 DNA. Immature monkey monocyte-derived DC were nucleofected with 10 μg gfp mRNA or pEGFP-N1 DNA, and GFP expression was determined at various intervals posttransfection by flow cytometry.

Relationship between DNA and mRNA nucleofection and DC maturation.Immature DC are specialized in antigen uptake, and as such, transfection of DC with either DNA or mRNA is traditionally done at the immature stage of differentiation. However, electroporation of mature human DC with mRNA is reported to be efficient (22, 31), and in the murine system, DNA transfection may be enhanced in mature DC (28). To determine the influence of DC maturation on transfection efficiency, we nucleofected human monocyte-derived DC with DNA and mRNA encoding gfp with and without prior treatment of cells with CD40L for 24 h to induce maturation. Maturation was confirmed by phenotypic analysis using flow cytometry (data not shown). Prior DC maturation did not enhance GFP expression, as nucleofection of DC with DNA or mRNA generated similar levels of transgene expression regardless of the maturation state at the time of gene delivery (Fig. 4a).

• Open in new tab
• Download powerpoint

FIG. 4.

Differential responsiveness of human DC nucleofected with gfp mRNA and pEGFP-N1 DNA to CD40L. (a) Human monocyte-derived DC were nucleofected with 10 μg pEGFP-N1 DNA or 10 μg gfp mRNA (solid line) or mock nucleofected (dotted line) with and without prior treatment of cells with 3 μg/ml CD40L as indicated for 24 h, and GFP expression was determined by flow cytometry. Numbers represent the percentage of cells expressing GFP based on mock transfection. (b and c) Immature human DC were nucleofected with DNA (b) (solid histograms) or mRNA (c) (solid histograms) or mock transfected (open histograms) and cultured with or without 3 μg/ml CD40L for 24 h as indicated. Cells were labeled with phycoerythrin-conjugated antibodies as indicated and analyzed by flow cytometry. Of the cells nucleofected with DNA or mRNA, only GFP-expressing DC are shown based on gating.

We next evaluated the responsiveness of DNA- and mRNA-nucleofected DC to maturation. Human immature DC were nucleofected with pEGFP-N1 DNA or gfp mRNA or were mock nucleofected and either left untreated or treated immediately with CD40L. Cells were harvested and their phenotype was analyzed 24 h postnucleofection by flow cytometry. To ensure that only the phenotype of transgene-expressing cells was analyzed, cells were gated based on GFP expression. Mock-nucleofected immature DC expressed relatively high levels of HLA-DR, CD86, and CD40 and lacked expression of CD80 and CD83, as expected (Fig. 4b and c) (6). Nucleofection of immature DC with pEGFP-N1 DNA did not induce maturation and in fact resulted in a modest downregulation in CD40 expression compared to that of mock-nucleofected cells (Fig. 4b). Moreover, immature DC nucleofected with DNA were refractory to CD40L, as GFP-expressing DC had negligible increases in CD83, CD86, and CD80 expression following CD40 ligation compared to those of mock-nucleofected controls (Fig. 4b). In contrast, immature DC nucleofected with gfp mRNA had a minor shift in expression of CD83 and CD80 compared to that of mock-nucleofected DC and responded fully to subsequent CD40 ligation, with increases in CD83, CD86, and CD80 expression, similar to those of mock-nucleofected cells (Fig. 4c). These data indicate that mRNA-nucleofected DC are responsive, whereas DNA-nucleofected DC are refractory to CD40L-mediated maturation.

mRNA- but not DNA-nucleofected DC stimulate robust virus-specific T-cell responses when expressing wild-type virus genes.A key issue in DC transfection for immunotherapy applications is the capacity to stimulate T-cell responses to the specific transgene being introduced. This has been evaluated in the past using lipofection and electroporation approaches to deliver DNA or mRNA into monocyte-derived DC (41, 51); however, the ability to stimulate T-cell responses following DNA and mRNA transfection of DC via nucleofection, which we have shown is a superior method of transfection, has not been compared. To evaluate this, we cultured monocyte-derived DC from a rhesus macaque that had a robust CD8⁺ T-cell response to SIV Gag through vaccination (33). Immature DC were nucleofected with pGag-N1 DNA encoding a SIVmac239 gag sequence isolated from an SIV-infected animal, or the corresponding gag mRNA (33), and immediately matured with CD40L for various intervals to induce maturation. To ensure applicability to immunotherapeutic vaccine strategies designed to stimulate T-cell responses to autologous viral sequences (48), we used mRNA and DNA encoding wild-type, non-codon-optimized gag sequences. The transfected DC were then cultured with autologous PBMC at a 1:10 ratio in a short-term IFN-γ ELISPOT assay to determine their capacity to stimulate Gag-specific effector T cells (33). DC that had been nucleofected with gag mRNA 24 h earlier induced robust Gag-specific T-cell responses with a frequency of greater than 5,000 IFN-γ spot-forming cells per million PBMC (Fig. 5). Importantly, DC that had been nucleofected with gag-expressing mRNA 48 h previously induced reduced but still strong T-cell responses, supporting the notion that antigen expressed following the mRNA delivery of transgene is processed and presented in DC for extended periods. In contrast to DC nucleofected with mRNA, DC nucleofected with DNA encoding wild-type gag generated no detectable responses above the nonspecific responses generated by DC nucleofected with p-EGFP-N1 DNA at either 24 h or 48 h postnucleofection (Fig. 5). These data indicate that DC nucleofected with wild-type viral mRNA have a durable presentation of antigen to T cells and stimulate robust virus-specific T-cell responses, whereas DC nucleofected with a comparable DNA construct are ineffective.

• Open in new tab
• Download powerpoint

FIG. 5.

Monkey DC nucleofected with non-codon-optimized mRNA but not with DNA encoding SIV gag stimulate potent Gag-specific effector T-cell responses. Immature DC propagated from an SIV Gag-vaccinated monkey were mock nucleofected (Mock) or nucleofected with DNA or mRNA encoding GFP or SIVmac239 Gag as indicated and matured with CD40L for 24 h (left) or 48 h (right) prior to incubation with autologous PBMC in an IFN-γ ELISPOT assay. IFN-γ spot-forming cells (SFC) were measured 24 h later. Data shown are means ± standard errors of the means (SEM) of triplicate determinations.