Antiviral protection after DNA vaccination is short lived and not enhanced by CpG DNA

Abstract

In this study, we investigated the potential of a DNA vaccine expressing the minimal cytotoxic T lymphocyte (CTL) epitope gp33 of the lymphocytic choriomeningitis virus glycoprotein to protect against infection of a non-lymphoid organ and compared this to protection against a systemic infection. Furthermore, since immune stimulatory sequences have been shown to augment CTL responses, we examined the capacity of CpG DNA to enhance CTL memory. The data show that DNA vaccination with a gp33-based gene construct induced short-lived gp33-specific CTL which protected against a systemic infection but not against a peripheral infection. Immune stimulatory sequences were incapable of either prolonging CTL memory or promoting protection against infection of a peripheral organ.

Figure 1

(a) The gp33 minimal epitope gene construct. The LCMV GP minimal epitope was flanked by three leucines and four alanines. For detection the FLAG tag was fused N-terminally and EGFP C-terminally to the minimal epitope. Expression of the fusion protein is driven by the cytomegalovirus (CMV) promoter. The expression plasmid was named pEGFPL33A. (b) Expression levels of the minimal epitope fusion protein. MBL-2 B-cell lymphoma cells were stably transfected with the fusion gene construct and expression levels of the minimal epitope gene product were determined by flow cytometry. One representative clone MBGF33 is shown. Non-transfected MBL-2 cells served as negative controls. (c) Specific lysis of MBGF33. Transfectants served as ⁵¹Cr-labelled target cells in a 5-hr cytotoxicity assay. LCMV-specific splenic effector CTL were obtained from mice which had been infected i.v. with 200 PFU of LCMV WE 8 days previously. To evaluate maximal antigen presentation and cytotoxicity the respective target cells were pulsed with gp33 peptide. Non-transfected MBL-2 cells served as controls. Spontaneous release was < 20% for all targets. A representative example of two independent experiments is shown.

Figure 2

(a, d) DNA vaccination induces gp33-specific CTL. C57BL/6 mice were immunized once with pEGFPL33A-coated gold beads (three discharges on the shaved abdomen) using a Helios gene gun device. After 10 days spleen cells were restimulated in vitro for 5 days in medium without (a) or with (d) 5% Con A supernatant by co-culturing with LCMV-infected macrophages. Standard cultures were serially diluted and specific cytotoxicity was tested in a 5-hr ⁵¹Cr release assay on gp33-pulsed or unpulsed EL-4 target cells. Spontaneous release was < 20%. (b, e) ISS do not enhance the generation of gp33-specific CTL. Mice were immunized as described above with gold beads coated with pEGFPL33A and ISS. After 10 days spleen cells were restimulated for 5 days in the absence (b) or presence (e) of Con A supernatant and cultures were tested for gp33-specific cytotoxicity as described for (a). LCMV-immune mice (200 PFU LCMV i.v. 30 days before restimulation) served as controls (c, f). For statistical analysis the data were subjected to a Mann–Whitney U-test.

Figure 3

Detection of ISS coupled to the gold beads. Coupled DNA was eluted from the gold beads and radioactively end-labelled with [γ-³²P]ATP using polynucleotide kinase. The labelled products were separated on a 19% polyacrylamide gel and detected in autoradiographs. In parallel the indicated amounts of ISS were end-labelled as controls.

Figure 4

(a) Recovery of gp33-specific memory CTL after DNA vaccination. C57BL/6 mice were immunized with pEGFPL33A or pEGFPL33A/ISS as described for Fig. 2. After 41 days mice were restimulated in vivo by i.v. challenge with 5 × 10⁵ PFU LCMV WE and 4 days thereafter spleen cells were assayed in a 5-hr ⁵¹Cr release assay. Since 4 days are not sufficient for effector CTL to develop from naive CTL, this read out system assays for memory CTL. EL-4 target cells were pulsed with gp33 peptide or were left unpulsed. Spontaneous release was < 20%. (b) Antiviral protection against systemic LCMV infection is short-lived after DNA vaccination. C57BL/6 mice were immunized with pEGFPL33A or pEGFPL33A/ISS as described for Fig. 2. LCMV infected (200 PFU LCMV i.v.) and naive mice served as controls. Ten or 34 days after immunization mice were challenged i.v. with 5 × 10⁵ PFU LCMV WE and 3 days later LCMV titres were determined in the spleen. For statistical analysis the data were subjected to a Mann–Whitney U-test.

Figure 5

DNA vaccination fails to induce protection against infection of a peripheral organ. C57BL/6 mice were immunized with pEGFPL33A or pEGFPL33A/ISS as described for Fig. 2. Ten days after immunization mice were challenged i.p. with 2 × 10⁶ PFU recombinant vaccinia virus expressing LCMV GP (Vacc-G2) or with irrelevant recombinant vaccinia virus (Vacc-VSV) as controls. LCMV-infected mice served as controls for virus clearance. Four days after the vaccinia challenge infection, vaccinia titres were measured in the ovaries.