A Comparison of Red Fluorescent Proteins to Model DNA Vaccine Expression by Whole Animal In Vivo Imaging

Abstract

DNA vaccines can be manufactured cheaply, easily and rapidly and have performed well in pre-clinical animal studies. However, clinical trials have so far been disappointing, failing to evoke a strong immune response, possibly due to poor antigen expression. To improve antigen expression, improved technology to monitor DNA vaccine transfection efficiency is required. In the current study, we compared plasmid encoded tdTomato, mCherry, Katushka, tdKatushka2 and luciferase as reporter proteins for whole animal in vivo imaging. The intramuscular, subcutaneous and tattooing routes were compared and electroporation was used to enhance expression. We observed that overall, fluorescent proteins were not a good tool to assess expression from DNA plasmids, with a highly heterogeneous response between animals. Of the proteins used, intramuscular delivery of DNA encoding either tdTomato or luciferase gave the clearest signal, with some Katushka and tdKatushka2 signal observed. Subcutaneous delivery was weakly visible and nothing was observed following DNA tattooing. DNA encoding haemagglutinin was used to determine whether immune responses mirrored visible expression levels. A protective immune response against H1N1 influenza was induced by all routes, even after a single dose of DNA, though qualitative differences were observed, with tattooing leading to high antibody responses and subcutaneous DNA leading to high CD8 responses. We conclude that of the reporter proteins used, expression from DNA plasmids can best be assessed using tdTomato or luciferase. But, the disconnect between visible expression level and immunogenicity suggests that in vivo whole animal imaging of fluorescent proteins has limited utility for predicting DNA vaccine efficacy.

Fig 1. Comparison of red fluorescent proteins in vitro.

CHO-k1 cells were transfected with 1μg DNA complexed with Lipofectamine. At various time points, cells were imaged by fluorescent microscopy—excitation 540–580nm and emission 600–660nm (a). Fluorescence intensity was measured by image analysis software at 550–600 nm (b) or 600–700 nm (c). Images in (a) representative of n = 3 experiments, points in b and c represent mean +/- SEM of 4 repeats, *** p<0.001 comparing tdTomato and other proteins in panel b, ## p<0.01 comparing tdKatushka2 and other proteins in panel c, by 2 way ANOVA.

Fig 2. Comparison of red fluorescent proteins after intramuscular DNA delivery.

Mice were injected i.m. with 50μg DNA in the anterior tibialis muscle and the site electroporated. At days 3, 5 and 7 after transfection, mice were imaged in vivo using an Imaging System at 550nm excitation and 600nm emission. Representative of n = 3 experiments. White circles indicate fluorescent protein expression.

Fig 3. Comparison of red fluorescent proteins after subcutaneous DNA delivery.

Mice were injected s.c. with 50μg DNA in the epigastric region and the site electroporated. At days 3, 5 and 7 after transfection, mice were imaged in vivo using an Imaging System at 550nm excitation and 600nm emission. Representative of n = 3 experiments. White circles indicate fluorescent protein expression.

Fig 4. Comparison of luciferase expression after DNA by various routes.

50μg DNA encoding luciferase was delivered by the i.m., s.c. or tattooing routes. At various time points after transfection, mice were injected i.p. with Luciferin substrate and imaged in vivo using an Imaging System. Representative images from day 7 after transfection by the i.m. (a), s.c. (b) and tattoo (c) routes. Relative expression levels were quantified using image analysis software (d), points represent mean +/- SEM of n = 5 animals.

Fig 5. Comparison of immunogenicity after DNA by various routes.

5μg DNA encoding H1 haemagglutinin was delivered by the i.m., s.c. or tattooing routes on days 0 and 14. On day 28, mice were infected with 5x10⁵ H1N1 influenza in 100μL intranasally. Weight loss was monitored following infection (a), serum HA1 specific IgG (b), and lung influenza specific (pentamer stained) CD8+ T cells (c) were measured on day 7 after infection. In a subsequent study, mice were immunised with 5μg DNA encoding H1 haemagglutinin delivered by the i.m., s.c. or tattooing routes on days 0 and were infected on day 14 with 5x10⁵ H1N1 influenza in 100μl intranasally. Weight loss was monitored following infection (d), serum HA1 specific IgG (e), and lung influenza specific (pentamer stained) CD8+ T cells (f) were measured on day 7 after infection. Points represent n = 5 animals +/- SEM (a) or individuals (b, c) +/- SEM. # p<0.05, ### p<0.001 between control and other groups analysed by 2 way ANOVA (a), * p<0.05, ** p<0.001 as indicated, analysed by one way ANOVA and post test (b, c).