• gynecology;
  • virology;
  • immunology;
  • gene shuffling;
  • cervical cancer;
  • immunotherapy;
  • tumor regression;
  • DNA vaccine;
  • genetic adjuvants


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Treatment of patients with cervical cancer by conventional methods (mainly surgery, but also radiotherapy and chemotherapy) results in a significant loss in quality of life. A therapeutic DNA vaccine directed to tumor-specific antigens of the human papilloma virus (HPV) could be an attractive treatment option. We have developed a nontransforming HPV-16 E7-based DNA vaccine containing all putative T cell epitopes (HPV-16 E7SH). DNA vaccines, however, are less immunogenic than protein- or peptide-based vaccines in larger animals and humans. In this study, we have investigated an adjuvant gene support of the HPV-16 E7SH therapeutic cervical cancer vaccine. DNA encoded cytokines (IL-2, IL-12, GM-CSF, IFN-γ) and the chemokine MIP1-α were co-applied either simultaneously or at different time points pre- or post-E7SH vaccination. In addition, sequence-optimized adjuvant genes were compared to wild type genes. Three combinations investigated lead to an enhanced IFN-γ response of the induced T cells in mice. Interestingly, IFN-γ secretion of splenocytes did not strictly correlate with tumor response in tumor regression experiments. Gene-encoded MIP-1α applied 5 days prior to E7SH-immunization combined with IFN-γ or IL-12 (3 days) or IL-2 (5 days) postimmunization lead to a significantly enhanced tumor response that was clearly associated with granzyme B secretion and target cells lysis. Our results suggest that a conditioning application and combination with adjuvant genes may be a promising strategy to enhance synergistically immune responses by DNA immunization for the treatment of cervical cancer. © 2009 UICC

Around 370,000 cases of cervical cancer are newly diagnosed each year worldwide and nearly 200,000 deaths are related to this disease.1 Moreover, in Third World countries cervical cancer is one of the major cause of cancer-associated deaths. Even under optimal circumstances of medical care 40% of patients with cervical cancer die of the disease.2 The human papilloma virus (HPV) high-risk type 16 (HPV-16) is responsible for about 50% of the cancers.3 The transforming activity of the high-risk HPV types has been mainly assigned to the oncoprotein E74 that interferes with the cell cycle control by interaction with pRb.5 Because the E7 is exclusively and consistently expressed by the HPV-infected tumor cells and all precancerous lesions it represents an ideal target for a tumor-specific immune therapy.

A new and very promising approach in vaccine development is the application of naked DNA. As compared to protein- or peptide-based vaccines a DNA vaccine has remarkable advantages making it of potential interest for Third World countries: (i) production costs are relatively low and predictable, (ii) DNA is stable and does not require refrigeration for storage, (iii) no unwanted immune reactions against other components of the vaccine are observed as it can be the case in vector-based vaccines, thus DNA vaccines can be used for repeated boosting6 and (iv) DNA is easy to modify and therefore adaptable to different needs.

Therefore, we have developed an artificial HPV-16 E7 gene (HPV-16 E7SH) that contains all natural occurring epitopes but lacks transforming properties. The E7SH gene proved to be immunogenic in mice and in vitro in human lymphocyte cultures as measured by IFN-γ Elispot-assay and 51Cr-release assays. Importantly, it demonstrated high effectivity in tumor regression experiments.7 Despite these encouraging results in the murine model, the main disadvantage of DNA vaccines is their weak immunogenicity as compared to protein- or peptide-based vaccines in larger animals,8 nonhuman primates9 and humans.10 To circumvent this disadvantage, efforts were recently undertaken to enhance effectiveness of DNA-based vaccines,11–13 making them attractive for the use in humans. One of the most promising approaches is the use of genetic adjuvants like cytokines and chemokines that are co-applied with the antigen.14 In principle, co-injection of DNA encoding for adjuvant cytokines may be a strategy to increase cellular immune responses to a DNA vaccine. The cytokines IL-2 and IFN-γ, for example, stimulate the Th1 immune response and thus the generation of cytotoxic T lymphocytes (CTLs).15 Also IL-12 expressed by B cells and macrophages is of particular importance because it favors a Th1-response.16, 17 GM-CSF (granulocyte-macrophage colony-stimulating factor) as a classical Th0 cytokine (also referred to as pro-inflammatory cytokine) is known to enhance general immunogenicitiy without altering the Th1/Th2 balance significantly. GM-CSF promotes the differentiation and activation of macrophages and dendritic cells (DCs).18 Chemokines mediate chemoattractive activity for specific types of leukocytes. Macrophage inflammatory protein 1α (MIP-1α), which binds to CC chemokine receptor 5 (CCR5), recruits immature DCs to the site of inoculation, resulting in enhanced induction of the cellular and humoral immune responses.19

Some investigators found that the application of a gene-encoded cytokine/chemokine before or after injection of the antigen-encoding plasmid is more effective than a concurrent treatment14, 20–22 or that combination of several genetic adjuvants is even more powerful.21, 23, 24 Until today only very limited data investigating the optimal conditions (type of genetic adjuvant, time-point of application, combination) are available. We have hypothesized that an optimal combination of adjuvant genes mediating recruitment, maturation and activation of lymphocytes should be concerted resulting in enhanced DNA vaccine potency. In this study, we have investigated the optimal combination and the optimal timing of application of the gene-encoded HPV-16 E7SH antigen and the most promising adjuvant genes for therapeutic settings (IL-2, IL-12, IFN-γ, GM-CSF, MIP-1α).

Material and methods

  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

Generation of the recombinant vectors

The HPV-16 E7SH gene was cloned via 5′HindIII and 3′XbaI into the pTHamp immunization vector,25 which has already been used in a version without any antibiotic resistance gene (pTHr) in humans26 as described earlier.7 Wildtype (WT) adjuvant genes (amplified from different plasmids received from NIH, USA) used were IL-2 WT, IL-12 WT, GM-CSF WT and IFN-γ WT. Amplified genes were cloned via 5′HindIII and 3′XbaI into the immunization vector pTHamp and sequenced.

The codon-optimized adjuvant genes IL-2 opt, IL-12 opt GM-CSF opt, IFN-γ opt and MIP-1α opt were generated in cooperation with GENEART (Regensburg, Germany). To enhance expression, codons were optimized for the human system (which is nearly identical to the murine system). Moreover, during the optimization process different cis-acting sequences (internal TATA-boxes, chi-sites and ribosomal entry sites, AT-rich or GC-rich (>80% or <30%) sequence stretches, ARE, INS, CRS sequence elements, repeat sequences and RNA secondary structures, (cryptic) splice donor and acceptor sites, branch points) were avoided. The synthetic genes were also cloned into pTHamp via 5′HindIII and 3′XbaI restriction sites into the pTHamp vector.

RT-PCR analysis

RNA from transfected NIH3T3 cells (Effectene Transfection Reagent, Qiagen, Hilden, Germany) was isolated by NucleoSpin total RNA isolation (Macherey-Nagel, Düren, Germany) according to the manufacturers protocol. For RT-PCR OneStep RT-PCR kit (Qiagen, Hilden, Germany) was used according to the user manual. Primers for RT-PCR were as follows: 5′ TAC AGC ATG GAG CTC GCA TCC 3′ (IL-2 WT forward), 5′ GTT CCT GTA ATT CTC CAT CCT 3′ (IL-2 WT reverse), 5′ CGC TGA CGC TCG TGC TGC TCG 3′ (IL-2 opt forward), 5′ ACT TGA ACG TCA GCA TCC GCG 3′ (IL-2 opt reverse), 5′ AAC GCT ACA CAC TGC ATC TTG 3′ (IFN-γ WT forward), 5′ GAT TTT CAT GTC ACC ATC CTT 3′ (IFN-γ WT reverse), 5′ CGC TGC AGC TGT TCC TGA TGG 3′ (IFN-γ opt forward), 5′ GCA GGT AGA ACG AGA TGA TCT 3′ (IFN-γ opt reverse), 5′ TGG CTG CAG AAT TTA CTT TTC 3′ (GM-CSF WT forward), 5′ TAG CTT CTT GAA GGA GAA CTC 3′ (GM-CSF WT reverse), 5′ TAT GTT GTA GAG GTG GAC TGG 3′ (IL-12 WT forward, p35), 5′ ATG TGA GTG GCT CAG AGT CTC 3′ (IL-12 WT reverse, p35), 5′ TGG CTG CAG AAC CTG CTG TTC 3′ (GM-CSF opt forward), 5′ TTC TTG AAC GAG AAC TCG TTC 3′ (GM-CSF opt reverse), 5′AAG GTG TCG ACG ACC GCG CTC 3′ (MIP-1α opt forward), 5′ GGC GCA GAT CTG CCG GTT CCG 3′ (MIP-1α opt reverse), 5′ TCG CGA CGC TCG CGC TGC TGA 3′ (IL-12 opt forward, p40), 5′ TCG ACG TCT GGT CGC GCG TGA 3′ (IL-12 opt reverse, p40). DNA was amplified for 30 min at 50°C (1 cycle), 15 min at 95 °C (1 cycle), [1 min at 94 °C, 1 min at 60°C, 1 min at 72°C (30 cycles)] and 10 min at 72°C (1 cycle).

Real-time PCR

NIH3T3 cells were co-transfected with the eGFP expression vector (pCMV-eGFP, Addgene) together with adjuvant-encoding vectors, respectively, using Effectene transfection reagent (Qiagen, Hilden, Germany) according to the manufacturer's instructions. Eighteen hours posttransfection we analyzed the cells on a FACScan™ flow cytometer (BD Biosciences). RNA from GFP-expressing cells was isolated by NucleoSpin total RNA isolation (Macherey-Nagel, Düren, Germany) according to the manufacturer's protocol. One microgram of pure total RNA was reverse transcribed using an oligo dT primer (Reverse Transcription System kit, Promega). Quantitative real-time RT-PCR was performed with the LightCycler® instrument (Roche, Mannheim, Germany) using the LightCycler® FastStart DNA Master SYBR Green I (Roche) reaction mix using primers as given earlier [10 sec at 94°C, 10 sec at 60°C and 8 sec at 72°C (40 cycles)]. Amplification of the correct product was confirmed by a melting point curve analysis followed by agarose gel electrophoresis.

Intracellular cytokine staining

Protein expression of cloned adjuvant genes was detected by intracellular staining. NIH3T3 cells were co-transfected with the eGFP expression vector (pCMV-eGFP, Addgene) together with adjuvant-encoding vectors, respectively, using Effectene transfection reagent (Qiagen, Hilden, Germany) according to the manufacturer's instructions. After 48 hr 2 × 106 cells were brefeldin A treated (10 μg/ml) for 4 hr. Following fixation with 4% paraformaldehyde at 4°C for 15 min, the cells were incubated for 30 min on ice in the dark with cytokine-specific antibody: rat anti-mouse FITC conjugated IL-2 (eBioscience, Cat. No. 11-7021, 20 μg/106 cells), rat anti-mouse FITC conjugated IFN-γ (Biomol, Cat. No. 17662-15M2, 10 μg/106 cells), rat anti-mouse PE-conjugated GM-CSF (IMGENEX, Cat. No. IMG-226D, 1 μg/106 cells), rat anti-mouse PE-conjugated IL-12 (p40) (eBioscience, Cat. No. 12-7123, 0.25 μg/106 cells) in PBS containing 2% FCS and 0.1% (w/v) Saponin (Sigma-Aldrich). Samples were washed twice and resuspended in 400 μl of FACS buffer prior to flow cytometry analysis. Adjuvant gene expression was verified by gating GFP-positive cells and expressed as mean fluorescence intensity (MFI).

Western blot analysis

Protein expression of cloned adjuvant genes was also detected by Western blot. NIH3T3 cells (1 × 106) were lysed 24 hr after transfection (Effectene Transfection Reagent, Qiagen, Hilden, Germany) by boiling for 10 min in SDS sample buffer and directly separated by SDS-PAGE. Recombinant E. coli expressed proteins (GM-CSF, ImmunoTools, Cat. No. 12343122; IL-2, Biomol, Cat. No. 17663-26A; IL-12, Biomol, Cat. No. 18434; IFN-γ, Chemicon, Cat. No. IF005; MIP-1α, Biomol, Cat. No. M1202-09A) were used as positive controls. The proteins were transferred to Immobilon-P transfer-membrane (Millipore, Bedford, USA) by a transblot semi-dry transfer cell (BioRad, München, Germany). To detect adjuvant proteins the following antibodies were used as follows: rat anti-mouse IL-2 (BD Pharmingen, 554424, 2 μg/ml), goat anti-mouse IL-12 (Sigma-Aldrich, 17642, 1 μg/ml), goat anti-mouse IFN-γ (Sigma-Aldrich, 19141, 0.2 μg/ml), rabbit anti-mouse MIP-1α (Biomol, M1202-26, 0.2 μg/ml), rat anti-mouse GM-CSF (IMGENEX, IMG-226E, 2 μg/ml). The following corresponding secondary antibodies were used: rabbit anti-goat HRP (Dako, P0449, 1:2,000), swine anti-rabbit HRP (Dako, P0399, 1:3,000), goat anti-rat HRP (Jackson Immuno Research, 112-035-003, 1:5,000). Anti β-actin antibody (Biomol, 905-733-100, 2 μg/ml) was used as a loading control.

DNA vaccination

Female C57BL/6 mice (own breed) were kept under SPF isolation conditions and standard diet at the animal facilities of the University of Constance, Germany. Agarose-gel verified plasmids (>95% supercoiled) (QIAGEN EndoFree Plasmid Kit; preparations contained less than 0.1 endotoxin units/μg plasmid DNA as tested earlier by Limulus endotoxin assay) were applied to 6- to 8-week-old female mice into each musculus tibialis (50 μl of plasmid DNA, 1 μg/μl in PBS). Ten to 12 days after vaccination animals were sacrificed and spleens were isolated. All operations on live animals were performed under Isofluran anesthesia (CuraMed Pharma, Karlsruhe, Germany). The institutional review board approved the study.

Cell lines and culture conditions

All cell lines used were of C57BL/6 origin (H2b context). RMA cells27 and RMA-E7 (HPV-16 E7 WT gene) transfectants 2F1128 were cultured in RPMI 1640 supplemented with heat-inactivated 5% (v/v) fetal calf serum (FCS, Gibco, Eggenstein, Germany), 2 mM L-glutamine, penicillin (100 U/ml) and streptomycin (100 μg/ml). In addition, G418 (0.8 mg/ml) was added to RMA-E7 (2F11) cultures to maintain E7 expression. C3 tumor cells derived from embryonic mouse cells transfected with the complete HPV-16 genome29 were cultured in the same medium, supplemented with kanamycin (0.1 mg/ml).

Splenocytes were cultured in αMEM (Sigma, Deisenhofen, Germany) supplemented with 10% FCS, 0.1 mM β-mercaptoethanol, 4 mM glutamine and antibiotics as above for the first 4–5 days after splenectomy. Subsequently, the spleen cells were cultured in αMEM+ supplemented with 2.5% supernatant of a concanavalin-A-induced rat spleen cell culture as a source of murine IL-2 and 25 mM methyl-α-mannopyranosid (Sigma, Deisenhofen, Germany).

In vitro restimulation of murine CTL lines

2 × 107 spleen cells (pretreated with ACT lysis buffer [17 mM Tris/HCl, 160 mM NH4Cl, pH 7.2] to deplete erythrocytes) were co-cultured with 2 × 106 irradiated (100 Gy) RMA (controls) or RMA-E7 2F11 cells in 25 cm2 culture flasks. First in vitro restimulations were performed at the day of the spleen isolation and were repeated weekly up to 4 times. Four to 5 days after the first in vitro restimulation, the spleen cell cultures were distributed into 24-well plates (every single culture was titrated over 6 wells) using 2 ml of αMEM+ medium per well. Beginning with the second in vitro restimulation additionally to the RMA/RMA-E7 cells (1 × 105 per well), irradiated (100 Gy) cells DC2.4 cells (kindly provided by K. Rock, University of Massachusetts Medical School, Worcester, MA) were added as a source of the co-stimulating B.7 molecule. Cultures were grown at 37°C and 7.5% CO2 in a humidified incubator.

IFN-γ/granzyme B Elispot assays

Murine IFN-γ Elispot assays were performed ex vivo and 5 or 6 days after each in vitro restimulation as described earlier.7 The granzyme B Elispot assay was performed accordingly to the IFN-γ Elispot Assay. The anti-mouse granzyme capture antibody (100 ng/well, clone R4-6A2; PharMingen, San Diego, CA) and as detection antibody the biotinylated anti-mouse granzyme antibody (50 ng/well, clone XMG1.2; PharMingen, San Diego, CA) were used. An animal was scored positive when the amount of IFN-γ secreting cells was at least 100% over the control animal (empty vector) showing the highest amount of IFN-γ secreting cells.

51Cr-release assays

The 51Cr-release assays were performed 5–6 days after an in vitro restimulation of murine spleen cells in parallel to the Elispot assays. 1 × 104 Na251CrO4 labeled (50 Ci) target cells/well (RMA or RMA-E7) were incubated together with decreasing numbers of effector cells in 200 μl per well of a 96-well round bottom plate (Costar, Corning, USA) for 4 hr. Subsequently, 50 μl of supernatant was harvested from each well and the released radioactivity was measured in a Microbeta counter (Wallac, Turku, Finland). Specific lysis was calculated according to the formula: percent specific lysis = [(cpm of the sample-spontaneous release)/(total release-spontaneous release)] × 100, where total release and spontaneous release are measured in counts per minute (cpm). Spontaneous chromium release was determined by using 51Cr-labeled target cells without effector cells, and total chromium release was determined by adding 2% Triton X-100 to lyse the labeled target cells. An animal was scored positive when the specific lysis of a specific target (RMA-E7 cells) was at least 20% above the lysis of the control (RMA cells). The variance between control wells was < 5%.

Tumor regression studies

C57BL/6 mice received 0.5 × 106 C3 cells in 100 μl of PBS subcutaneously in the right shaved flank. When small tumors were palpable in all animals (days 5–18) the first DNA-injection (recombinant or control plasmid) was applied i.m. in both musculus tibialis anterior. The boost-vaccination was performed 12–15 days later. Tumor sizes were measured with a caliper and were determined every 2–4 days until mice had to be sacrificed (tumor size of 400 mm2 or when tumors were bleeding). Tumor sizes of the mice within a group were calculated as arithmetic means with standard error of the means (SEMs). In the tumor regression experiments an individual was counted as “regressor,” when the tumor area at the endpoint of each experiment was within the “0–25 mm2” field.

Statistical analysis

Differences of means between experimental and control group were considered statistically significant when p was <0.05 by unpaired Students t-test.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

The aim of this study was to enhance cellular immune response by DNA vaccination. We investigated the co-application and time-shifted application of different genetic adjuvants (IL-2, IL-12, IFN-γ, GM-CSF, MIP-1α) in conjunction with a DNA-encoded model antigen (HPV-16 E7SH). A further aspect of our study was, to proof the potential of DNA-optimization by (i) codon optimization, (ii) low GC content and (iii) depletion of negative cis-acting sequences of the genetic adjuvants.

The HPV-16 E7 WT gene acts as viral oncogene encoding an oncoprotein that binds the pRb (retinoblastoma) tumor suppressor protein.5 Therefore, it cannot be used unmodified in humans. As a consequence, we have generated an artificial (“shuffled”) HPV-16 E7-gene (HPV-16 E7SH) containing all putative CTL epitopes and exhibiting high safety features. We showed the induction of a strong E7-wildtype (E7WT) directed cellular and humoral immune response including tumor protection and regression after in vivo immunization in the murine system. Moreover, the vaccine demonstrated immunogenicity in humans, demonstrated by priming of antigen-specific T cells in vitro. Importantly, the artificial HPV-16 E7 gene has completely lost its transforming properties as measured in soft agar transformation assays.7 In this study we used the artificial HPV-16 E7SH gene as a model antigen.

Effects of co-application of WT cytokine genes on IFN-γ secretion and tumor regression

Initially, we tried to enhance the cellular immune response by co-application of adjuvant WT genes IFN-γ, IL-2, IL-12 or GM-CSF together with the gene encoded model antigen HPV-16 E7SH (E7SH). For all 4 cytokines positive effects had been shown in the context of DNA vaccination.15, 30, 31 Immune responses were determined in 3 independently performed immunization experiments by IFN-γ Elispot assay. A total of 12 mice per group were immunized at the same time together with 100 μg E7SH DNA and 100 μg plasmid encoding for the respective adjuvant gene (E7SH/IFN-γ WT, E7SH/IL-2 WT, E7SH/GM-CSF WT, E7SH/IL-12 WT). One group (n = 12) received 100 μg E7SH DNA only, control animals (n = 12) received empty vector. Control animals were used to define the cut off and Elispots were set as background. An animal was scored positive when the number of spots was at least 2-fold above the highest number of IFN-γ secreting cells of the control animals. To our surprise, no difference in the number of positive evaluated animals and no enhancement of the frequency of IFN-γ secreting cells per 2 × 104 splenocytes were observed in the E7SH/IL-2 group (9/12 mice; 36 ± 8) versus the E7SH group (8/12 mice; 32 ± 6). All other combinations led to a decreased IFN-γ response compared to the E7SH only group. In line with the absolute number of IFN-γ secreting cells in all other groups less animals were found positive (data not shown). This result is in clear contrast to the findings of other groups.15, 30, 31

We investigated the same plasmid combinations after 2 immunizations (days 0 and 12–15) in 3 independent tumor regression experiments. When the first tumor had reached a mean size of 400 mm2 (or when the first tumor started bleeding) the experiments were terminated (days 35–48). In contrast to the findings in the IFN-γ Elispot assays, we observed a not quite significantly (p = 0.06) retarded tumor growth pattern in the E7SH/IL-2 WT group versus the E7SH control group (28 ± 8 mm2vs. 48 ± 6 mm2, respectively; vector control: 100 ± 32 mm2). None of the animals showed total tumor regression. The other combinations with adjuvant genes investigated showed similar tumor growth rates as in the E7SH control group (data not shown). Taken together, only the IL-2 WT gene was able to enhance tumor regression although without detectable enhancement of IFN-γ secreting T cells. Probably, this finding is a hint that not only IFN-γ secreting splenocytes contribute to the slower tumor growth in the E7SH/IL-2 WT group.

Sequence optimized adjuvant genes are stronger expressed in comparison to their WT counterparts

In an effort to enhance expression of adjuvant genes, we improved translation by codon optimization for the human system that is nearly identical to the murine system. Expression of the generated vectors (termed pTHamp-“adjuvant gene” opt) was verified by RT-PCR and real time PCR (data not shown), intracellular cytokine staining and by Western blotting. For quantification of WT and sequence-optimized adjuvant genes by real time PCR we performed co-transfection of NIH3T3 cells together with a GFP-encoding expression vector. GFP positive cells were used to determine transfection efficiencies and mRNA levels were measured by real time RT-PCR. We found that mRNA levels of all sequence-optimized genes were about 2-fold higher as compared to their WT counterparts (data not shown). Next, we demonstrated relative expression rate by Western blotting. For this purpose, murine NIH3T3 fibroblasts were transiently transfected with the WT or optimized adjuvant genes. Much higher protein expression of the sequence-optimized genes was found. The expression (as judged by naked eye) could be estimated to be between 3-fold (IL-12) and 15-fold (GM-CSF) higher (Fig. 1a). We also determined adjuvant gene expression by intracellular staining of eGFP/adjuvant gene co-transfected NIH3T3 cells (using short-term brefeldin A treatment as protein transport inhibitor). In all cells transfected with sequence-optimized genes an enhancement of mean fluorescence (MFI) of gated double positive cells (eGFP+/adjuvant gene) was consistently observed in 3 independent experiments (Fig. 1b). Expression enhancement was strongest for GM-CSF (14-fold), followed by IFN-γ and IL-2 (10-fold, respectively) and IL-12 (4-fold) that corresponds well to the data obtained by the Western blotting experiments. Therefore, we decided to use optimized adjuvant genes in the following experiments.

thumbnail image

Figure 1. Enhancement of expression strength of adjuvant genes by sequence optimization. Expression of WT sequence versus optimized (opt) sequence adjuvant genes IFN-γ, GM-CSF, IL-12, IL-2 after transfection of NIH3T3 cells. (a) Evaluation was done by Western blot. Sixty micrograms of total protein from transfected NIH3T3 cells were loaded. As a loading control β-actin was used. Recombinant protein of IFN-γ, GM-CSF, IL-12 and IL-2 were used as positive controls, respectively (not shown). The expression experiment was performed twice with comparable results. (b) Intracellular cytokine staining after co-transfection with eGFP. Fluorescence intensity of double positive cells is expressed as mean fluorescence intensity (MFI). The expression experiment was performed twice with comparable results.

Download figure to PowerPoint

Co-application of optimized adjuvant genes is not generally able to enhance IFN-γ secretion of splenocytes

To determine if optimized adjuvant genes were able to induce a stronger cellular immune response in comparison to WT counterparts, we compared WT genes of IFN-γ, IL-2, GM-CSF and IL-12 with their sequence optimized version. A total of 12 mice per group (E7SH only, E7SH/IFN-γ, E7SH/IL-2, E7SH/GM-CSF, E7SH/IL-12) were immunized once with 100 μg DNA of the E7SH-vector together with the same amount of adjuvant gene encoding plasmid. The control animals (in total n = 12) received empty vector (100 μg/animal). Again we did not find a clear increase in the frequency of IFN-γ secreting cells after co-application of optimized adjuvant genes. The number of positively reacting mice in the optimized genetic adjuvant groups in comparison to WT was moderately enhanced. This was seen for E7SH/IFN-γ (6/12 WT vs. 9/12 opt) and for E7SH/IL-2 (3/12 WT vs. 5/12 opt) groups, but not in the others (data not shown). The absolute numbers of IFN-γ secreting cells, however, were not significantly higher in the E7SH/IFN-γ and E7SH/IL-2 groups (21 ± 6 WT vs. 35 ± 5 opt and 11 ± 7 WT vs. 16 ± 4 opt/2 × 104 splenocytes, respectively). The other adjuvant genes did not show any difference due to gene optimization (data not shown). In contrast, Elispot assays performed for granzyme B revealed a 2- to 3-fold higher response in all animals receiving optimized genes (Table I). Statistically highly significant results were found for IFN-γ (p: 0.004), IL-2 (p: 0.003) and IL-12 (p: 0.002). From these, data we conclude that by gene optimization not necessarily stronger IFN-γ secretion but consistently enhanced granzyme B secretion could be achieved.

Table I. Granzyme B Response after Co-Application of Sequence Optimized Adjuvant Genes
DNA construct combination used for vaccinationMean of granzyme B secreting cells/ 2 × 104 splenocytes ± SEM1,2
WT genetic adjuvants Opt genetic adjuvants
  • 1

    Four mice per group immunized i.m. in 3 independent experiments with 100 μg empty vector (control) or with pTHamp vector encoding E7SH and/or pTHamp vector encoding wildtype sequence (WT) or sequence optimized (opt) adjuvant genes IFN-γ, GM-CSF, IL-2 or IL-12, respectively. One representative of 3 independently performed experiments is shown.

  • 2

    After 1st in vitro restimulation.

Control4 ± 2
E7SH only33 ± 5
E7SH/IFN-γ37 ± 6 114 ± 17
E7SH/IL-242 ± 5 93 ± 11
E7SH/GM-CSF31 ± 7 58 ± 12
E7SH/IL-1233 ± 4 87 ± 9

Gene encoded MIP-1α enhances IFN-γ secretion when delivered 5 days prior to an antigen encoding plasmid

Recombinant MIP-1α has been shown to recruit DCs to the site of vaccine inoculation in mice, leading to enhanced induction of the cellular and humoral immune responses.19 We therefore injected MIP-1α gene encoding vector (100 μg/animal) 5 days prior to or at the same time as E7SH antigen (100 μg / animal) or at the same time point as the E7SH gene (total n = 8/group). Controls received empty vector, E7SH only or MIP-1α alone (total n = 8, each). Both empty vector and MIP-1α treated mice showed no E7-specific IFN-γ secretion in Elispot assays. MIP-1α given at the same time point as the antigen (day 0) resulted in no change in cellular immune responses compared to the E7SH group. Mice receiving the chemokine 5 days prior to E7SH DNA had a significantly higher IFN-γ response (p: 0.007) in Elispot assays (Fig. 2). The same finding was true for granyzme B secretion (data not shown). This experiment suggests that it is of advantage to recruit DCs to the site of antigen application prior to providing the antigen and that this is possible by application of a plasmid encoded MIP-1α.

thumbnail image

Figure 2. Ex vivo IFN-γ Elispot responses after DNA immunization. Four mice per group were immunized i.m. with 100 μg empty vector or with pTHamp vector encoding MIP-1α and/or E7SH. Each bar represents the number of activated T cells from an individual animal. The number gives the mean ± SEM. One representative of 2 experiments is shown.

Download figure to PowerPoint

An application of optimized adjuvant genes at different time points is advantageous to enhance IFN-γ secretion

Because of the lack of clear immunostimulatory effects by co-application of optimized adjuvant genes, we took advantage of the findings by others that for some cytokines the timing of administration is critical. For GM-CSF the application of the cytokine-encoded DNA 3 days after DNA vaccination has been shown to induce the strongest Th1-response.21 In the same study the co-administration of IL-12 and GM-CSF at the same time point led to a significantly improved DTH response in mice. For IL-2 the application 5 days after DNA vaccination resulted in highest IFN-γ and IL-2 secretion by splenocytes.20 Therefore, we next investigated the effects of different adjuvant gene combinations together with MIP-1α pretreatment 5 days before E7SH DNA injection. We hypothesized that an increased immunogenicity of the model antigen encoding plasmid when applied 5 days after gene encoded MIP-1α results from the recruitment of DCs at the site of injection. Therefore, we tried to enhance the cellular immune responses synergistically by combining DNA-based cytokines associated with DC maturation (e.g. IFN-γ) and activation (e.g. GM-CSF) at different time points (see also discussion). First, we determined if MIP-1α enhances immune responses also in the context of a time-shifted application regimen. Four mice per group were immunized in 2 independently performed settings (total: n = 8/group, once 100 μg/plasmid/animal) with or without pretreatment of MIP-1α (day-5)/E7SH (day 0) and IL-2 opt either at day 0 or +5. In IFN-γ Elispot assays we found strongest responses in animals receiving a MIP-1α pretreatment before and the cytokine gene 5 days after immunization, followed by the MIP-1α (day-5)/E7SH (day 0) and IL-2 (day 0) group collective (Table II). Because MIP-1α DNA seemed to be beneficial when given 5 days prior gene-encoded antigen we generally used this pretreatment in the following experiments.

Table II. Effects of Site Conditioning by Time-Shifted Application of Sequence Optimized Adjuvant Genes
Application of sequence optimized adjuvant genes at different time points1Mean of IFN-γ secreting cells/2 × 104 splenocytes ± SEM
Experiment I2Experiment II2Summary (experiments I + II)
  • 1

    Four mice per group immunized i.m. in 2 independent experiments with 100 μg empty vector (control) or with pTHamp vector encoding E7SH and/or pTHamp vector encoding sequence optimized adjuvant genes MIP-1α, IFN-γ, GM-CSF, IL-2 or IL-12, respectively.

  • 2

    After 1st in vitro restimulation.

Control2 ± 24 ± 23 ± 2
E7SH only29 ± 539 ± 534 ± 5
MIP-1α (−5)/E7SH (0)50 ± 1758 ± 1054 ± 14
MIP-1α (−5)/E7SH (0)/IL-2 (0)67 ± 1177 ± 772 ± 9
MIP-1α (−5)/E7SH (0)/IL-2 (5)131 ± 25113 ± 18122 ± 22
MIP-1α (−5)/E7SH (0)/GM-CSF (0)28 ± 634 ± 1231 ± 9
MIP-1α (−5)/E7SH (0)/GM-CSF (3)36 ± 522 ± 529 ± 10
MIP-1α (−5)/E7SH (0)/IFN-γ (0)39 ± 1133 ± 736 ± 9
MIP-1α (−5)/E7SH (0)/IFN-γ (3)70 ± 2584 ± 2277 ± 24
MIP-1α (−5)/E7SH (0)/IL-12 (0)38 ± 1028 ± 1233 ± 11
MIP-1α (−5)/E7SH (0)/IL-12 (3)35 ± 1421 ± 428 ± 9
MIP-1α (−5)/E7SH (0)/IL-12 + GM-CSF (3)65 ± 2079 ± 2472 ± 22

Again, animals were injected (in total: n = 8 mice/group) once with MIP-1α (day-5), E7SH (day 0) and adjuvant genes at different time points as indicated (100 μg/plasmid/animal), respectively. No clear enhancement in the absolute number of IFN-γ secreting cells was observed between the E7SH only versus GM-CSF (day 0), GM-CSF (day 3), IFN-γ (day 0), IL-12 (day 0) and IL-12 (day 3) groups (Table II). As compared to the E7SH only group (34 ±5 IFN-γ secreting cells/2 × 104 splenocytes), IFN-γ (day 3) (frequency of 77 ± 24) and IL-12+GM-CSF (day 3) (frequency 72 ± 22) treatment induced enhanced but statistically not significantly different cellular immune responses. For IL-2 (day 5) of 2 × 104 splenocytes 122 ± 22 IFN-γ secreting cells T-cells were seen (statistically highly significant, p: 0.0016).

These data suggest that IL-2 (given 5 days post immunization), IFN-γ (given 3 days post immunization) and IL-12+GM-CSF (given 3 days post immunization) increases IFN-γ secretion of splenocytes after MIP-1α pretreatment.

IFN-γ secretion of antigen-specific splenocytes does not obligatory correlate with tumor regression

Therapeutic tumor vaccines aim at induction of an effective immune response able to eradicate established tumors. Therefore, we investigated in parallel to the experiments given in Table II the same plasmid combinations in tumor regression experiments. Animals (n = 10/group/experiment) received a tumorigenic dose of syngeneic C3 cells that were transfected with the whole HPV-16 genome. When small tumors were palpable in all individuals and after MIP-1α pretreatment (day-5) animals were immunized with the HPV-16 E7SH encoding plasmid (day 0) and received DNA encoding adjuvant cytokines at optimal time points thereafter, as indicated. A boost-immunization was given 12–15 days after the prime, thereby repeating the same application regimen including the MIP-1α pretreatment. It was necessary to end the tumor regression experiment at day 44 when the tumor size of the first animals of the control group (empty vector at days −5, 0, 5) reached 400 mm2. At this time point 8 total regressors (no tumor mass palpable) were observed in the IFN-γ (day 3) and 5 regressors in the IL-2 (day 5) group. GM-CSF (day 3) treatment resulted in only 2 and E7SH control in 1 regressor. The absolute number of total regressors corresponded well with the final tumor size at day 44. In the IFN-γ (day 3) group the average tumor size of the 10 animals was 3 ± 3 mm2, while animals receiving IL-2 (day 5) and GM-CSF (day 3) had tumor sizes of 18 ± 8 mm2 and 49 ± 10 mm2, respectively. The control group had developed a mean tumor size of 365 ± 12 mm2 and the E7SH only group had 81 ± 17 (Table III). The therapeutic effect of the IL-12 (day 3) was comparable to IL-2 (day 5) with regard to the average tumor sizes at day 44, although no total regressors were observed in the IL-12 group (9/10 animals had tumor sizes up to 25 mm2 at day 44). No other combination did improve tumor responses to the E7SH vaccine (data not shown). A therapeutic effect was observed by all combinations listed in Table III, and differences of control treatment to IFN-γ- (p: 0.003), IL-2- (p: 0.004) and IL-12 (p: 0.004) groups were statistically significant. These data suggest that DNA vaccination combined with treatment in a time-shifted application of MIP-1α with IFN-γ, IL-2, IL-12 or GM-CSF at optimized time points can lead to a strong enhancement of tumor-specific immune responses.

Table III. Growth of Established C3 Tumors in Mice after Immunization With HPV-16 E7SH DNA ± Concurrent Application of Sequence Optimized Genetic Adjuvants
Tumor regression with sequence optimized adjuvant genes + MIP-1α pretreatment1Empty vectorE7SH DNAMIP-1α (−5), E7SH (0)MIP-1α (−5), E7SH (0), IFN-γ (3)MIP-1α (−5), E7SH (0), IL-2 (5)MIP-1α (−5), E7SH (0), IL-12 (3)MIP-1α (−5), E7SH (0), GM-CSF (3)
  • 1

    All combinations given in Table II were investigated in tumor regression experiments, the 4 combinations with strongest response are shown here.

  • 2

    Tumors were established when palpable and size was measured over time.

  • 3

    Experiment had to be terminated at day 44 due to tumors sizes of the control animals. One representative of 2 tumor regression experiments is shown.

Average tumor size at day 0 (mm2) ± S.E.M211 ± 212 ± 312 ± 411 ± 311 ± 111 ± 210 ± 2
Average tumor size at day 44 (mm2) ± S.E.M365 ± 1281 ± 1770 ± 163 ± 318 ± 818 ± 849 ± 10
Distribution of tumor sizes within the group at day 44 (mm2: number of animals)30–25: 00–25: 4 (1 total)0–25: 40–25: 10 (8 total)0–25: 8 (5 total)0–25: 90–25: 4 (2 total)
26–99: 026–99: 426–99: 526–99: 026–99: 226–99: 126–99: 6
100–199: 0100–199: 2100–199: 1100–199: 0100–199: 0100–199: 0100–199: 0
300–400: 10300–400: 0300–400: 0300–400: 0300–400: 0300–400: 0300–400: 0

Cellular immune response correlates with tumor regression

Interestingly, we did not observe a strong correlation of IFN-γ secretion (Table II) with tumor regression (Table III). For example, IL-12 (day 3) adjuvant gene-treated animals were able to mediate strong tumor regression, but failed to enhance IFN-γ secretion in comparison to the E7SH only control group. Therefore, we decided to address whether the ability to lyse target cells in vitro corresponds to tumor regression in vivo. For this purpose, splenocytes of tumor regressors were tested in 51Cr-release assays. Total tumor regressors (n = 3, respectively) were used of IFN-γ (day 3), IL-2 (day 5) and GM-CSF (day 3) groups, whereas for IL-12 (day 3) treatment 3 animals of the 0–25 mm2 group were used (smallest available tumor size, 2–5 mm2). From the E7SH group 1 total regressor and 2 animals with tumor sizes of 6 and 18 mm2 were used. After 2 rounds of in vitro restimulation, all tumor regressor animals of the adjuvanted groups (n = 3/group) displayed strong E7WT-specific lysis of RMA-E7 target cells (Fig. 3). Strongest specific lysis of E7-expressing cells was observed in the IFN-γ (day 3) group (73% ± 2%), which corresponds well to strongest tumor regression (Table III and Fig. 3). The mean of the specific lysis of the RMA-E7 cells in the GM-CSF (day 3) group was somewhat lower (56% ± 11%), followed by IL-2 (day 5) and IL-12 (day 3) (52% ± 4% and 48% ± 2%, respectively). In the E7SH only group 2 of 3 animals were scored positive (20% ± 4%). The splenocytes of the control animals (naïve mice) did not show any E7-specific lysis, demonstrating that E7-specific priming was induced in vivo. Together, these results show that tumor regression observed after E7SH immunization is associated with lysis of E7-expressing target cells in vitro but not obligatory with detection of significant IFN-γ secretion in Elispot assays.

thumbnail image

Figure 3. CTL responses in animals with tumor regression. Splenocytes of tumor regressors of the experiment shown in Table III (3 responders in each group) were tested by 51Cr-release assays after 2 rounds of in vitro restimulations for lysis of syngeneic parental RMA (diamonds) or E7-WT expressing RMA-E7 transfectants (triangles). Data gives the mean ± SEM.

Download figure to PowerPoint

To verify the correlation between IFN-γ secretion, granzyme B secretion and specific cell lysis after adjuvant gene-supported DNA immunization in detail, we investigated tumor regressor derived splenocytes for their ability to secrete IFN-γ and granzyme B in Elispot assays. In summary, 14 of 15 tumor-regressor derived splenocytes were able to lyse target cells in vitro (as measured by 51Cr-release assay). T cells of 13 of 15 of these individuals also secreted granzyme B. This correlation suggests that cytotoxic CD8 T cells mediate cell lysis also in vivo in tumor responders. Moreover, contrary to this finding only 7/15 mice were scored positive in IFN-γ Elispot assays (Table IV). Therefore, IFN-γ secretion not necessarily indicates tumor regression and measuring IFN-γ response may not be the best surrogate parameter of antitumor responses after immunizations. Interestingly, we observed a correlation between the absolute number of granzyme B secreting cells, percentage of specific lysis of target cells in vitro and the number of animals positively evaluated for tumor regression. For example, the strongest tumor response was observed within the IFN-γ (day 3) group––splenocytes of these responders mediated highest granzyme B secretion and specific lysis in vitro (Tables III and IV). Although we tested only a limited number of animals in this set of experiments, we conclude that tumor regression may be strongly associated with granzyme B secretion of splenocytes but not necessarily with IFN-γ secretion.

Table IV. Sensitivity of Read Out Assays IFN-γ Elispot Versus Granzyme B Elispot Versus51Cr-Release Assay for Detection of CTL Activity in Animals With Tumor Regression
Reactivity of tumor regressor derived splenocytesCTL-responses in animals with tumor regression1 (mean of IFN-γ or granzyme B secreting cells/2 × 104 splenocytes ± SEM or specific lysis in %, [number of animals positively evaluated2])
IFN-γ Elispot3Granzyme B Elispot351Cr-relase assay4
  • 1

    Splenocytes of tumor regressors (tumor size 0–25 mm2) of the experiment shown in Table III (3 of each group) were used.

  • 2

    Control animals were used to define the cut off and ELISpots/unspecific lysis of control targets were set as background. An animal was scored positive when the number of IFN-γ/granzyme B secreting cells was at least 2-fold above the highest number of IFN-γ/granzyme B secreting cells of the control animals (empty vector recipients) and the specific lysis of the specific target (RMA-E7 cells) was at least 20% above the lysis of the control target (RMA cells).

  • 3

    Ex vivo.

  • 4

    After 2nd in vitro restimulation, mean ± SEM of the maximal specific lysis (RMA-E7 cells) of the 3 animals, respectively.

Control23 ± 1 [0/3]4 ± 2 [0/3]7 ± 1 [0/3]
E7SH only17 ± 14 [1/3]79 ± 36 [2/3]20 ± 4 [2/3]
MIP-1α (−5)/E7SH (0)/IFN-γ (3)10 ± 6 [1/3]229 ± 16 [3/3]73 ± 2 [3/3]
MIP-1α (−5)/E7SH (0)/IL-2 (5)19 ± 9 [2/3]119 ± 48 [3/3]52 ± 4 [3/3]
MIP-1α (−5)/E7SH (0)/IL-12 (3)6 ± 3 [1/3]192 ± 8 [3/3]48 ± 2 [3/3]
MIP-1α (−5)/E7SH (0)/GM-CSF (3)22 ± 12 [2/3]94 ± 58 [2/3]56 ± 11 [3/3]

Together, these experiments imply that an optimally scheduled application of optimized sequences of adjuvant genes has the potential to significantly enhance cellular immune responses to DNA vaccines as demonstrated here in vitro by granzyme B Elispot assays and 51Cr-release assays and in vivo in tumor regression experiments.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We have demonstrated that supportive cytokines are able to enhance in vitro granzyme B secretion of splenocytes and the in vivo tumor response after DNA immunization in mice. A time-shifted application of vaccination site conditioning and T cell supportive cytokines given as sequence-optimized genes in expression vectors enhanced cellular immunity in response to a DNA vaccine significantly.

Naked DNA vaccines have become very promising in the field of vaccinology. They are attractive alternatives to expensive and unstable protein- and peptide-based vaccines. Indeed, a number of clinical trials have been performed with naked DNA without serious side effects.10, 32, 33 Unfortunately, DNA-based vaccines are relatively weakly immunogenic in large animals and primates8–10, 33 compared to protein-based vaccines. Therefore, there is an urgent need for development of more advanced and highly immunogenic DNA vaccines.

Efforts were undertaken to enhance immunogenicity of DNA,11–13 one of the most auspicious is the combination with adjuvants, like cytokines and chemokines.14 These molecules are able to recruit DCs to the application site (in most cases the muscle) and to activate them. During conversion of immature to mature DCs, a number of changes are observed, e.g. increased expression of MHC and co-stimulatory molecules and secretion of cytokines and chemokines for attraction and expansion of activated T cells. Using conventional adjuvants a Th1 or Th2 biased immune response can be triggered. Indeed, the clinical use of different recombinant cytokines/chemokines is well-established, however, high costs of purification, short half life resulting in a need of multiple injections and systemic toxicity limits their practical use.34 Therefore, several investigators have used gene-encoded cytokines co-applied with the plasmid DNA encoding for the antigen.

In a first approach we tried to enhance immunogenicity of our experimental therapeutic HPV-16 E7SH gene7 by co-application with each 1 of 4 selected WT DNA-encoded cytokines. All 4 cytokines, namely IL-2, IL-12, IFN-γ and GM-CSF favor therapeutic immune responses and for each an enhancement of the cellular immune system has been documented.15, 30, 31 To our surprise, the simultaneous administration of WT IFN-γ, GM-CSF and IL-12 expressing vectors with the E7SH expressing plasmid did not significantly alter the pattern of immune responses as measured by IFN-γ Elispot assays and by tumor regression experiments. Only the IL-2 gene enhanced IFN-γ secretion of splenocytes and slowed down tumor growth. This unexpected finding is in contrast to others,15, 30, 31 who used the same cytokines/chemokine but described different effects on the immune system regarding the Th1/Th2 ratio. This may be due to differences in the route and time of application making direct comparisons difficult. We have performed experiments 3 times independently with expression proofed plasmids by RT-PCR and real time PCR (data not shown), by intracellular cytokine staining and Western blot with the same outcome. One could hypothesize that the expression vectors used in previous reports were superior leading to stronger expression. The plasmid pTH, however, used in this study is approved for the use in humans26 and therefore we decided not to change the plasmid backbone.

To enhance gene expression, we tried gene sequence optimization (codon-usage) and additionally avoided extreme GC-contents and negatively cis-acting sequences where possible. Real-time PCR (data not shown) revealed a 2-fold increase of mRNA in the case of sequence-optimized genes. Because we have not manipulated promoter sequences, we speculate that this observation is rather a consequence of increased mRNA-stability due to the optimization process introduced by GENART than of enhanced translation efficiency. By intracellular cytokine staining as well as by Western blotting stronger protein expression was detected in all sequence-optimized genes investigated and, moreover, the expression levels corresponded well in both detection systems.

In fact site conditioning with optimized IFN-γ and IL-2 augmented IFN-γ secretion of induced T cell responses as compared to WT genes. Importantly, when granzyme B secretion was used as a read out assay and compared to the IFN-γ Elispot results, we observed a clearly positive effect by all sequence optimized cytokines. The IFN-γ Elispot assay is widely accepted as a suitable surrogate marker for the detection of cytotoxic T cells and is often used for immunological monitoring of cancer vaccines. It may, however, underestimate specific T cell frequency while the granzyme B assay directly measures the release of a cytolytic protein upon effector-target interaction.

In initial immunization experiments we used cardiotoxin site pretreatment and obtained a total tumor regression in 6 of 10 animals.7 However, cardiotoxin is not allowed for use in humans. In this study, we omitted any cardiotoxin pretreatment resulting in a somewhat higher total average tumor size and complete regression in only 1 individual. To compensate for cardiotoxin and to further enhance DNA vaccine immunogenicity we included the gene-encoded chemokine MIP-1α (macrophage inflammatory protein 1α, a β-chemokine) in our studies. MIP-1α binds to CC chemokine receptor 5 (CCR5) on immature DCs and recruits these pAPCs to the site of inoculation, resulting in enhanced induction of the cellular and humoral immune responses.19 In our hands pretreatment with gene-encoded MIP-1α 5 days prior to immunization enhanced T cell responses (IFN-γ and granzyme B secretion) while co-application did not. In line with our results, Kim et al.35 found strong enhancement of antibody responses but not of CTL responses after co-application of MIP-1α together with a model antigen. Others also found a time-shifted application of chemokines prior immunization superior for induction of CTLs.36 However, some investigators found effects also after a co-application regimen.37 Our data suggest that the site conditioning with MIP-1α may support availability of DCs at the site of inoculation and may be a critical factor for the induction of CTLs after DNA vaccination.

At least for some cytokines the timing of administration has been shown to be important. For GM-CSF the application of the cytokine-encoded DNA 3 days after DNA vaccination has been shown to induce the strongest Th1-response.21 In the same study, the co-administration of IL-12 and GM-CSF at the same time point led to a significantly improved DTH response in mice. For IL-2 the application 5 days after DNA vaccination resulted in highest IFN-γ and IL-2 secretion by splenocytes.20 By combining the different approaches for site-conditioning, vaccination and post-delivery of cytokines, we demonstrated strongest cellular immune response in mice receiving MIP-1α pretreatment (day −5) prior to E7SH immunization (day 0) followed by IL-2 application 5 days later (day 5). Therefore, we investigated by CTL analysis and tumor regression experiments the most promising combinations of sequence optimized adjuvant genes and their application at optimized time points after MIP-1α pretreatment. We selected GM-CSF due to its effects on differentiation and activation of attracted DCs and IL-2, IL-12 and IFN-γ due to their ability to induce a Th1-shift. In addition, we combined GM-CSF and IL-12 treatment with the aim to combine DC differentiation and activation in a cytokine environment favoring induction of a cellular immune response.

Now, we found strongly enhanced cellular immune responses as measured in initial IFN-γ Elispot assays in mice receiving IL-2 (day 5) or IFN-γ (day 3). Tumor regression studies performed in parallel revealed the strongest effects in the IFN-γ (day 3), IL-2 (day 5) and IL-12 (day 3) treated groups. Because of the discrepancy between the IFN-γ Elispot data and the tumor regression experiments, we concluded that IFN-γ secretion as measured in vitro in Elispot assays does not correlate necessarily with cytotoxic effects in vivo. We therefore performed IFN-γ and granzyme B Elispot assays as well as 51Cr-release assays with tumor regressor derived splenocytes and found a strong correlation between the secretion of granzyme B and the lysis of target cells in 51Cr-release assays, but not as well with IFN-γ secretion. In the limited number of animals investigated in this setting, the granzyme B Elispot assay corresponds to the 51Cr-release assay for monitoring of a cell-mediated cytotoxicity and reflects the actual induction of CTLs with an ability to lyse tumor cells in vivo. The importance of granzyme B secretion in the context of therapeutic settings is also highlighted by the fact that granzyme B-deficient mice possess a strongly reduced antiviral response and tumor cell clearance,38 but an unchanged IFN-γ response.

It should be mentioned that the experiments shown were performed ex vivo or after first in vitro restimulation (Elispot assays) and after 2 rounds of in vitro restimulation (less sensitive 51Cr-release assay). The fact that we needed in some experiments none but in others 1 in vitro restimulation to detect specific T cells is probably due to differences in the immunization procedure. For example, we cannot exclude the possibility that some DNA solution leaked out from the muscle after injection reducing the amount of vaccine and therefore the strength of the induced response. Moreover, it is known that ccc-supercoiled (covalently closed, circular) DNA can change topography in samples stored at −20°C. Because a high take-up rate of muscle cells after DNA injection may depend on the ccc-topography this could also be a reason for the observed variations in our study. However, the control animals (empty vector treated) had only low background levels in all experiments shown demonstrating that E7-specific responses observed in vitro had actually been primed in vivo.

We are the first to show that the combination of a gene-encoded chemokine treatment at different time points presents a possible strategy to modulate synergistically the induction of specific immunity by DNA vaccination. In mice MIP-1α pretreatment (day −5) in combination with sequence-optimized IFN-γ (day 3) resulted in strongest responses in tumor regression studies. Here, we have demonstrated that 2 quite different separate functions of genetic adjuvants, the pre and postconditioning of the vaccination site can be combined. This may be a strategy to overcome DNA vaccine limitations and may have important implications for designing DNA vaccine strategies to treat cancer as well as infectious diseases. However, the future development of this promising approach will depend on the validation with more antigens and a crucial issue will be to proof the effectiveness in larger animals and in human beings.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References

We thank Prof. Dr. Marcus Groettrup (Universität Konstanz, Konstanz, Germany) for providing his lab infrastructure and GENEART (Regensburg, Germany) for cooperation in the production of sequence-optimized genes and Stefan Heymel for technical support.


  1. Top of page
  2. Abstract
  3. Material and methods
  4. Results
  5. Discussion
  6. Acknowledgements
  7. References
  • 1
    Pisani S,Gallinelli C,Seganti L,Lukic A,Nobili F,Vetrano G,Imperi M,Degener AM,Chiarini F. Detection of viral and bacterial infections in women with normal and abnormal colposcopy. Eur J Gynaecol Oncol 1999; 20: 6973.
  • 2
    Gatta G,Lasota MB,Verdecchia A. Survival of European women with gynaecological tumours, during the period 1978–1989. EUROCARE Working. Group. Eur J Cancer 1998; 34: 221825.
  • 3
    Bosch FX,Manos MM,Munoz N,Sherman M,Jansen AM,Peto J,Schiffman MH,Moreno V,Kurman R,Shah KV. Prevalence of human papillomavirus in cervical cancer: a worldwide perspective.International biological study on cervical cancer (IBSCC) Study. Group. J Natl Cancer Inst 1995; 87: 796802.
  • 4
    Dyson N,Howley PM,Munger K,Harlow E. The human papilloma virus-16 E7 oncoprotein is able to bind to the retinoblastoma gene product. Science 1989; 243: 9347.
  • 5
    Munger K,Scheffner M,Huibregtse JM,Howley PM. Interactions of HPV E6 and E7 oncoproteins with tumour suppressor gene products. Cancer Surv 1992; 12: 197217.
  • 6
    Liu MA. Vaccine developments. Nat Med 1998; 4: 51519.
  • 7
    Ohlschlager P,Pes M,Osen W,Durst M,Schneider A,Gissmann L,Kaufmann AM. An improved rearranged human papillomavirus type 16 E7 DNA vaccine candidate (HPV-16 E7SH) induces an E7 wildtype-specific T cell response. Vaccine 2006; 24: 288093.
  • 8
    Babiuk LA,van Drunen Littel-van den H,Babiuk SL. Immunization of animals: from DNA to the dinner plate. Vet Immunol Immunopathol 1999; 72: 189202.
  • 9
    Egan MA,Charini WA,Kuroda MJ,Schmitz JE,Racz P,Tenner-Racz K,Manson K,Wyand M,Lifton MA,Nickerson CE,Fu T,Shiver JW, et al. Simian immunodeficiency virus (SIV) gag DNA-vaccinated rhesus monkeys develop secondary cytotoxic T-lymphocyte responses and control viral replication after pathogenic SIV infection. J Virol 2000; 74: 748595.
  • 10
    Wang R,Doolan DL,Le TP,Hedstrom RC,Coonan KM,Charoenvit Y,Jones TR,Hobart P,Margalith M,Ng J,Weiss WR,Sedegah M, et al. Induction of antigen-specific cytotoxic T lymphocytes in humans by a malaria DNA vaccine. Science 1998; 282: 47680.
  • 11
    Jechlinger W. Optimization and delivery of plasmid DNA for vaccination. Expert Rev Vaccines 2006; 5: 80325.
  • 12
    Daudel D,Weidinger G,Spreng S. Use of attenuated bacteria as delivery vectors for DNA vaccines. Expert Rev Vaccines 2007; 6: 97110.
  • 13
    Tsen SW,Paik AH,Hung CF,Wu TC. Enhancing DNA vaccine potency by modifying the properties of antigen-presenting cells. Expert Rev Vaccines 2007; 6: 22739.
  • 14
    Calarota SA,Weiner DB. Enhancement of human immunodeficiency virus type 1-DNA vaccine potency through incorporation of T-helper 1 molecular adjuvants. Immunol Rev 2004; 199: 8499.
  • 15
    Chow YH,Chiang BL,Lee YL,Chi WK,Lin WC,Chen YT,Tao MH. Development of Th1 and Th2 populations and the nature of immune responses to hepatitis B virus DNA vaccines can be modulated by codelivery of various cytokine genes. J Immunol 1998; 160: 13209.
  • 16
    Guler ML,Gorham JD,Dietrich WF,Murphy TL,Steen RG,Parvin CA,Fenoglio D,Grupe A,Peltz G,Murphy KM. Tpm1, a locus controlling IL-12 responsiveness, acts by a cell-autonomous mechanism. J Immunol 1999; 162: 133947.
  • 17
    Afonso LC,Scharton TM,Vieira LQ,Wysocka M,Trinchieri G,Scott P. The adjuvant effect of interleukin-12 in a vaccine against Leishmania major. Science 1994; 263: 2357.
  • 18
    Morrissey PJ,Bressler L,Park LS,Alpert A,Gillis S. Granulocyte-macrophage colony-stimulating factor augments the primary antibody response by enhancing the function of antigen-presenting cells. J Immunol 1987; 139: 111319.
  • 19
    McKay PF,Barouch DH,Santra S,Sumida SM,Jackson SS,Gorgone DA,Lifton MA,Letvin NL. Recruitment of different subsets of antigen-presenting cells selectively modulates DNA vaccine-elicited CD4+ and CD8+ T lymphocyte responses. Eur J Immunol 2004; 34: 101120.
  • 20
    Barouch DH,Santra S,Steenbeke TD,Zheng XX,Perry HC,Davies ME,Freed DC,Craiu A,Strom TB,Shiver JW,Letvin NL. Augmentation and suppression of immune responses to an HIV-1 DNA vaccine by plasmid cytokine/Ig administration. J Immunol 1998; 161: 187582.
  • 21
    Kusakabe K,Xin KQ,Katoh H,Sumino K,Hagiwara E,Kawamoto S,Okuda K,Miyagi Y,Aoki I,Nishioka K,Klinman D. The timing of GM-CSF expression plasmid administration influences the Th1/Th2 response induced by an HIV-1-specific DNA vaccine. J Immunol 2000; 164: 310211.
  • 22
    Liu MA,Wahren B,Karlsson Hedestam GB. DNA vaccines: recent developments and future possibilities. Hum Gene Ther 2006; 17: 105161.
  • 23
    Iwasaki A,Stiernholm BJ,Chan AK,Berinstein NL,Barber BH. Enhanced CTL responses mediated by plasmid DNA immunogens encoding costimulatory molecules and cytokines. J Immunol 1997; 158: 4591601.
  • 24
    Okada E,Sasaki S,Ishii N,Aoki I,Yasuda T,Nishioka K,Fukushima J,Miyazaki J,Wahren B,Okuda K. Intranasal immunization of a DNA vaccine with IL-12- and granulocyte-macrophage colony-stimulating factor (GM-CSF)-expressing plasmids in liposomes induces strong mucosal and cell-mediated immune responses against HIV-1 antigens. J Immunol 1997; 159: 363847.
  • 25
    Hanke T,Schneider J,Gilbert SC,Hill AV,McMichael A. DNA multi-CTL epitope vaccines for HIV and Plasmodium falciparum: immunogenicity in mice. Vaccine 1998; 16: 42635.
  • 26
    Hanke T,McMichael AJ. Design and construction of an experimental HIV-1 vaccine for a year-2000 clinical trial in Kenya. Nat Med 2000; 6: 9515.
  • 27
    Ljunggren HG,Karre K. Host resistance directed selectively against H-2-deficient lymphoma variants. Analysis of the mechanism. J Exp Med 1985; 162: 174559.
  • 28
    Speidel K,Osen W,Faath S,Hilgert I,Obst R,Braspenning J,Momburg F,Hammerling GJ,Rammensee HG. Priming of cytotoxic T lymphocytes by five heat-aggregated antigens in vivo: conditions, efficiency, and relation to antibody responses. Eur J Immunol 1997; 27: 23919.
  • 29
    Feltkamp MC,Smits HL,Vierboom MP,Minnaar RP,de Jongh BM,Drijfhout JW,ter Schegget J,Melief CJ,Kast WM. Vaccination with cytotoxic T lymphocyte epitope-containing peptide protects against a tumor induced by human papillomavirus type 16-transformed cells. Eur J Immunol 1993; 23: 22429.
  • 30
    Kim JJ,Trivedi NN,Nottingham LK,Morrison L,Tsai A,Hu Y,Mahalingam S,Dang K,Ahn L,Doyle NK,Wilson DM,Chattergoon MA, et al. Modulation of amplitude and direction of in vivo immune responses by co-administration of cytokine gene expression cassettes with DNA immunogens. Eur J Immunol 1998; 28: 1089103.
  • 31
    Sin JI,Kim JJ,Ugen KE,Ciccarelli RB,Higgins TJ,Weiner DB. Enhancement of protective humoral (Th2) and cell-mediated (Th1) immune responses against herpes simplex virus-2 through co-delivery of granulocyte-macrophage colony-stimulating factor expression cassettes. Eur J Immunol 1998; 28: 353040.
  • 32
    MacGregor RR,Boyer JD,Ugen KE,Lacy KE,Gluckman SJ,Bagarazzi ML,Chattergoon MA,Baine Y,Higgins TJ,Ciccarelli RB,Coney LR,Ginsberg RS, et al. First human trial of a DNA-based vaccine for treatment of human immunodeficiency virus type 1 infection: safety and host response. J Infect Dis 1998; 178: 92100.
  • 33
    Le TP,Coonan KM,Hedstrom RC,Charoenvit Y,Sedegah M,Epstein JE,Kumar S,Wang R,Doolan DL,Maguire JD,Parker SE,Hobart P, et al. Safety, tolerability and humoral immune responses after intramuscular administration of a malaria DNA vaccine to healthy adult volunteers. Vaccine 2000; 18: 1893901.
  • 34
    Jin P,Wang E,Provenzano M,Stroncek D,Marincola FM. Gene expression signatures of interleukin-2 in vivo and in vitro and their relation to anticancer therapy. Crit Rev Immunol 2007; 27: 43748.
  • 35
    Kim JJ,Yang JS,Dentchev T,Dang K,Weiner DB. Chemokine gene adjuvants can modulate immune responses induced by DNA vaccines. J Interferon Cytokine Res 2000; 20: 48798.
  • 36
    Song R,Liu S,Leong KW. Effects of MIP-1 alpha, MIP-3 alpha, and MIP-3 beta on the induction of HIV Gag-specific immune response with DNA vaccines. Mol Ther 2007; 15: 100715.
  • 37
    Shao W,Li X,Li M,Shi L,Qin Y,Li K. Macrophage inflammatory protein-1alpha expression plasmid enhances DNA vaccine-induced immune responses against HSV-2. Immunol Cell Biol 2005; 83: 62631.
    Direct Link:
  • 38
    Revell PA,Grossman WJ,Thomas DA,Cao X,Behl R,Ratner JA,Lu ZH,Ley TJ. B and the downstream granzymes C and/or F are important for cytotoxic lymphocyte functions. J Immunol 2005; 174: 212431.