Experiments with C57BL/6 and BALB/c mice (Beijing Huafukang Biology Technology Co., Ltd., China) were approved by the Ethical Committee for the Use of Laboratory Animals at Jilin University and were carried out in strict accordance with the National Institutes of Health Guide for the Care and Use of Laboratory Animals. Mice were female, 6 to 8 weeks old.

Mouse colorectal cancer cell line CT26 was preserved at the National Engineering Laboratory for AIDS Vaccine of Jilin University. Mouse lung cancer cell line Lewis (No.: GF123) was purchased from Shanghai Gefan Biotechnology Co., Ltd. Lewis-GFP-MS and CT26-GFP-MS cell lines stably expressing tumor antigens MUC1 and survivin were previously generated in our laboratory (31).

The DNA vaccine CpDV-IL2-sPD1/MS and rAd, Ad-MS, were previously constructed (31, 33). DNA vaccine plasmids were amplified and purified by Changchun BCHT Biotechnology Co. The rAd vaccine was manufactured by Shenzhen Yuanxing Gene Technology Co., Ltd., and contained 2 × 10¹¹ vp/ml. Both vaccines were diluted to the final concentration in PBS (phosphate-buffered saline) buffer.

Two immunization strategies were used in this research based on vaccination intervals, namely, a traditional immune strategy (normal immunization strategy) with 1- or 2-week intervals, and a fast immunization strategy (days 0–2–5) with 2-day and 3-day intervals. Both DNA and rAd vaccines were administered intramuscularly. DNA vaccine is vaccinated at a dosage of 100 µg or a triple dosage.

For tumor inoculation, C57BL/6 mice were subcutaneously injected with 1.3~1.4 × 10⁵ Lewis-GFP-MS cells in the right hind flank, and the day of inoculation was recorded as day 0. In the normal immunization strategy group, the DNA vaccine was administered on day 1. In the fast immunization group, the DNA vaccine was administered when the tumor reached approximately 50~100 mm³, which occurred between day 4 and day 8. Tumor size was measured using a caliper, and the tumor volume was calculated (1/2 length × width²). The average volume in control group mice on the day of treatment initiation was set to a value of 100% (34, 35). Tumor-bearing mice were euthanized when the maximum diameter measured over 1.5 cm.

Total RNA of muscle or inguinal lymph nodes samples was isolated using TRIzol reagent (Invitrogen). cDNA was synthesized, and qRT-PCR was then performed to analyze the expression of chemokine genes as previously described. The quantitative real-time PCR (qRT-PCR) primer sequences were as follows: β-actin forward 5′-CAAGCAGGAGTACGACGAGT-3′, reverse 5′-GGCTGGCATGAGGTGTGTA-3′; Ccl3 forward 5′-CCTCTGTCACCTGCTCAACA-3′, reverse 5′-GATGAATTGGCGTGGAATCT-3′; Ccl4 forward 5′-CCCACTTCCTGCTGTTTCTC-3′, reverse 5′-GAGGAGGCCTCTCCTGAAGT-3′; Ccl5 forward 5′-TGCCCACGTCAAGGAGTATTTC-3′, reverse 5′-AACCCACTTCTTCTCTGGG TTG-3′; Ccl19 forward 5′-CACTCACTCTCTGTGGCCT-3′, reverse 5′-GGGCCAGAGTGATTCACATC-3′; Ccl20 forward 5′-TGCTCTTCCTTGCTTTGGCATGGGTA-3′, reverse 5′-TCTGTGCAGTGATGTGCAGGTGAAGC-3′; and Ccl21 forward 5′-TACTGGGCTATCCTCTTGA-3′, reverse 5′-ATGGCTCAGATGATGACTCT-3′.

The spleen was dissected, washed with sterile PBS buffer, cut into pieces, ground, and filtered with gauze twice to remove connective tissue. Red blood cells were removed using red blood cell lysis buffer (ACK, BioLegend, San Diego, CA, USA). Cellular debris were removed via centrifugation. The concentration of splenocyte suspensions was adjusted to 1 × 10⁷ cells/ml.

Tumor or muscle cell suspensions were prepared in a similar manner. However, prior to tissue grinding, tumor or muscle tissues were cut into pieces and incubated with collagenase I (final concentration: 40 μg/ml, Roche, Basel, Switzerland) for 2 h at 37°C and 5% CO₂. The tumor tissues were subjected to the same procedures as described for splenocyte suspensions.

IFN-γ released by splenocytes following MUC1- and survivin-specific stimulation was detected using an ELISpot kit as previously described. The peptide mixture, including survivin (H2-K^d sequence for BALB/c mice: AFLSVKKQF and H2-D^b sequence for C57BL/6 mice: STFKNWPFL, final concentration: 10 μg/ml) and MUC1 (H2-K^d sequence: APDTRPAPG and H2-D^b sequence: SAPDTRPAP, final concentration: 10 μg/ml), was used to stimulate splenocytes.

We used carboxyfluorescein diacetate succinimidyl ester (CFSE) staining to evaluate splenocyte cytotoxicity. Lewis-GFP-MS cells and wild-type Lewis cells were stained with high (5 μM) and low (0.5 μM) concentrations of CFSE. Splenocytes from C57BL/6 mice and the target Lewis-GFP-MS/Lewis cells were incubated at different effector-to-target (E:T) ratios for 8~10 h at 37°C and 5% CO₂, followed by analysis of dead target cell percentage among Lewis-GFP-MS cell and wild-type Lewis cell populations using a FACS Caliber instrument (BD Biosciences). Splenocytes from BALB/c mice were incubated with CT26-GFP-MS/CT26 treated with CFSE as described above. Lewis-GFP-MS/Lewis or CT26-GFP-MS/CT26 was determined by the strains of mice. Cytotoxicity activity was detected by flow cytometry, and specific lysis was calculated as follows: specific lysis (%) = [1 - (peptide-loaded cells/unloaded cells from the experiment group)/(peptide-loaded cells/unloaded cells from target cell control group)] × 100 (36, 37).

CD11c⁺ cells were isolated from different tissues using the CD11c MicroBeads UltraPure mouse kit (Miltenyi Biotec, Order no. 130-108-338) with an LS-positive selection column, as per the manufacturer’s instructions. Muscle tissues were cut into pieces and incubated with collagenase for 2 h at 37°C and 5% CO₂. A total of 10⁸ cells were resuspended in 400 µl buffer. CD11c Microbeads UltraPure (100 µl) were then added and incubated for 10 min in the dark at 2°C-8°C, followed by separation using the appropriate columns.

The frequency of different immune cell types was determined using FACS. Myeloid-derived suppressor cells (MDSCs) in tumor or spleen tissue were analyzed using a mouse MDSC Flow Cocktail 1 (CD11b PE/Gr-1 APC/Ly-6G FITC, BioLegend). Regulatory T cells were analyzed using the Mouse Regulatory T Cell Staining Kit #1 (eBioscience). FITC anti-mouse CD8α antibody (clone: 53-6.7, BioLegend), APC anti-mouse CCR5 antibody (clone: HM-CCR5, BioLegend), PE anti-mouse CCR6 antibody (clone: 29-2L17, BioLegend), and PE/Cy7 anti-mouse CCR7 antibody (clone: 4B12, BioLegend) were used to label isolated CD11c⁺ cells (38).

All statistical analyses were performed with GraphPad Prism software (version 7.0, GraphPad, Inc., La Jolla, CA, USA) using unpaired t-tests and One-way ANOVA, as previously described.

APC migration is driven by chemokines. To define which specific chemokines were upregulated after the first vaccination, we analyzed chemokine mRNA expression. The upregulation of a series of chemokines was observed in muscle or the ipsilateral inguinal lymph nodes throughout 0–168 h (time points: 0, 3, 6, 9, 12, 24, 48, 72, 96, and 168 h) after DNA vaccine administration ( Figure 1A ). The mRNA levels of Ccl3/Ccl4/Ccl5/Ccl20 (recruiting immature DCs, iDCs) and Ccl 19/Ccl 21 (recruiting mature DCs, mDCs) were determined ( Figure 1 ). At the injection site, Ccl3 was upregulated approximately 3.88~4.11-fold during 24~48 h post-vaccination, Ccl5 expression increased 1.57-fold at 48 h post-vaccination, and Ccl4 temporarily increased about 8-fold at 3 h post-vaccination ( Figure 1B ). Ccl19 increased 4.56-fold at 24 h, while Ccl21 increased 2.14~4.56 fold during 24~48 h ( Figure 1B ). When APCs uptake antigen at the injection site, they could carry the vaccine or antigen to lymph node acting the call for CCL19 or CCL21 (15, 26, 27). Hence, these chemokines mRNA were also analyzed in lymph nodes. Within the ipsilateral inguinal lymph nodes, of iDC-recruiting chemokines Ccl3/Ccl4/Ccl5/Ccl20, only Ccl20 exhibited a modest 1.294-fold increase at the 72-h time point, while low expression was observed for the rest ( Figure 1C ). mDC-recruiting Ccl21 was considerably upregulated 12.21-fold at the 9-h point, 9.30-fold at the 48-h point, 10.24-fold at the 72-h point, and 4.38-fold at the 96-h point ( Figure 1C ). These results suggested that local proinflammatory chemokines were upregulated at the injection site after DNA vaccination, which could enhance the recruitment of APCs, particularly DCs, and Ccl21 began to be upregulated in lymph nodes since 24 h later DNA vaccination.

CCR5 is the receptor of CCL3, CCL4, and CCL5, CCR6 is the receptor of CCL20, and CCR7 is the receptor of CCL19 and CCL20 (39). CCR5 and CCR6 could express on immature DCs and macrophage, interact with CCL3, CCL4, CCL5, and CCL20, and migrate APCs to the inflammatory site, including DNA vaccine injection site. When APCs internalize the vaccine, APCs downregulate CCR5 and CCR6, upregulate CCR7, and chemotactically migrate to lymph nodes by CCL19 and CCL21. Antigen presentation by APCs is a key process for immune response induction. CD11c is a prevalent APC marker (40, 41). To validate whether chemokines could recruit APCs into the injection site, CD11c⁺ cells were isolated from the injection site and ipsilateral inguinal lymph nodes at different time points after the first or second vaccination and analyzed via FACS. BALB/c mice were divided into four groups, and the immunization process is shown in Figure 2A . Two days after the first vaccination, we observed a nearly twofold increase of CD8⁺CD11c⁺ cell accumulation at the injection site. Five days post-vaccination, the number of CD8⁺CD11c⁺ cells decreased back to that of the PBS group. Upon the second vaccination 2 days after the first, CD8⁺CD11c⁺ cells accumulated back to previously observed levels after 3 days ( Figure 2B ). These results indicated that CD11c⁺ APCs were recruited to the injection site post-vaccination.

Chemokine receptor expression was also analyzed to determine CD11c⁺ cell migration. CD8⁺CCR5⁺CD11c⁺ cells, responding to CCL3, CCL5, and CCL20, exhibited a similar migration pattern to that of CD8⁺CD11c⁺ cells and were significantly recruited to the injection site post-vaccination (p < 0.05) ( Figure 2C ). CD8⁺CCR6⁺CD11c⁺ cells, also responding to CCL3, CCL5, and CCL20, were observed at the injection site after the second vaccination (p < 0.05) ( Figure 2D ). In addition, CCR7⁺ cells responding to CCL21, a marker of mDCs, exhibited a modest increase (p > 0.05) 2 days after the first vaccination, but no significant differences were observed between the four groups ( Figure 2E ).

CD11c⁺ cells migrated away from the injection site 5 days after the last vaccination in each group. Thereafter, we investigated whether these cells were attracted to the ipsilateral inguinal gland of the injection site. However, only CD8⁺CD11c⁺ cells appeared significantly increased two days after first injections (p < 0.05) ( Figure 2F ). iDCs expressing CCR5 or CCR6 exhibited a decline in ipsilateral inguinal gland ( Figures 2G, H ). No change was observed for CCR7⁺CD11c⁺ mDCs ( Figure 2I ).

Data in Figure 2 indicate that CD11c⁺ APCs may accumulate at the injection site 2 days after the first vaccination as well as after the administration of booster vaccines.

Based on the timing of CD11c⁺ cell accumulation at the injection site ( Figures 1 , 2 ), we speculated that vaccination at days 0, 2, and 5 would induce a more rapid and pronounced immune response. To assess this, healthy BALB/c mice were divided into three groups and vaccinated following fast or conventional strategies and sacrificed to then evaluate antitumor immune responses ( Figure 3A ). According to our previous results, administration of DNA vaccine CpDV-IL2-sPD1/MS under the normal immunization strategy could only induce a modest immune response. Further, rAd could boost this immune response. Thus, we questioned whether the rAd boost could further enhance the immune response under fast vaccination. Hence, for this experiment, both normal and fast DNA vaccination were boosted with 2-week-interval rAd vaccination. To analyze the antigen-specific cytolysis ability of splenocytes (effector cells), cancer cells CT26 with or without antigens (targets cells) labeled with different concentrations of CFSE were co-incubated with splenocytes with a ratio (effector cells: targets cells) of 50:1 or 100:1. FACS was used to detect the antigen-specific lysis. Cytotoxic T lymphocyte (CTL) assay results revealed that splenocytes from the fast vaccination group killed 50% more tumor cells than those of the normal vaccination group (p < 0.01, Figure 3B ). Further, the secretion of IFN-γ was 4~5-fold greater in the fast vaccination group compared with that in the normal vaccination group (p < 0.001, Figure 3C ).

We designed the fast vaccination strategy based on the timing of APC migration. Fast vaccination induced a stronger immune response than normal vaccination ( Figure 3 ). Moreover, as short-interval vaccination may cause prolonged antigen expression at the injection site, we questioned whether the fast vaccination strategy could also enhance the immune response via the accumulation of DNA vaccine-encoded antigens at the injection site.

To verify this conjecture, BALB/c mice were separated into a normal-dosage group, wherein mice were vaccinated using the fast vaccination strategy, and a triple-dosage group ( Figure 4A ). We employed CTL assay to analyze the antigen-specific cytolysis ability of splenocytes. At the 100:1 ratio of E:T, cytolysis of the fast vaccination group (11.83%) is higher than that of the triple-dose group (7.88%, p < 0.01) and that of the control group (5.61%, p <0.001). At 50:1 and 25:1 ratios, they have the same trend with the 100:1 ratio ( Figure 4B ). In ELISpot assay, splenocytes in the fast vaccination group secreted almost two-fold IFN-γ (353.5) of splenocytes in the triple-dose group (186) (p < 0.01, Figure 4C ). According to CTL and ELISpot assay results, DNA vaccine administration under the fast immunization strategy elicited a stronger immune response than did the triple-dosage strategy (p < 0.01) ( Figures 4B, C ). The percentages of CD69⁺ in CD8⁺ cells and MDSCs (immunosuppressive cells) among splenocytes indicated that triple dosage increased MDSCs and decreased effector cells (p < 0.001) ( Figures 4D, E ). Further, triple dosage did not enhance the immune response and induced immune suppression instead. Thus, we aimed to confirm whether the amount of antigen was indeed associated with immune responses. Hence, the DNA vaccine dosage was reduced to 50 or 25 μg, administered under the fast immunization strategy, for comparison with the 100-μg dosage based on the resulting immune response ( Figure 4F ). The CTL assay revealed no significant differences between 100 and 50 μg (p > 0.05) vaccination. Both the 100- and 50-μg vaccine dosages induced greater CTL-mediated cytolysis than did the 25-μg dosage at an E/T ratio of 100:1 (p < 0.01, Figure 4G ). ELISpot assay results confirmed no significant differences among the 100-, 50-, and 25-μg groups (p > 0.05, Figure 4H ). These results suggested that dosage does not play an important role in the fast vaccination strategy, as opposed to the traditional strategy, wherein the immune response was dose-dependent ( Figure S1 ).

The DNA vaccine dosage had little effect on the efficacy of the fast vaccination strategy as per the data presented in Figure 4 . Triple dosage at once or a lower dose administered on three separate occasions elicited slightly different or similar immune responses. This strongly suggests that the migration of APCs to the injection site played the key role under fast vaccination.

The above-described results highlighted the potential of fast vaccination as an alternative to the currently employed cancer vaccine administration regimens. As a new strategy, the expansion and contraction timeline of effector CD8⁺ T cells should be elucidated, and it would be useful when this fast DNA vaccination strategy is boosted with rAD vaccine in the following research. C57BL/6 mice were inoculated with Lewis-GFP-MS, vaccinated, and then sacrificed for analysis during the period of day 0 and day 14 after the last DNA vaccination ( Figure 5A ). To comprehensively reflect the antitumor effect, the tumor size was evaluated ( Figure 5B ). Three mice per group were sacrificed at each time point, and their tumors were dissected and weighed. When vaccines were administered under the fast immunization strategy, tumor growth was suppressed by 83.6% (p < 0.05, day 3), 72.86% (p < 0.05, day 7), 71.10% (p < 0.01, day 10), and 46.34% (p > 0.05, day 14) ( Figure 5B ). The function of antigen-specific IFN-γ secretion and cytolysis were also analyzed by ELISpot assay and CTL assay. ELISpot ( Figure 5C ) and CTL ( Figure 5D ) assays indicated that fast vaccination induced an effective immune response on the third day after the last vaccination, which persisted to the 10th day.

Tumor-infiltrating CD8⁺ T cells are thought to be the direct executor of killing tumor cells. Hence, we also analyzed the activated CD8⁺ T cells (CD8⁺/CD69⁺ cells) in both spleen ( Figure 5E ) and tumor tissue ( Figure 5F ) by FACS. At 7 and 14 days after the last vaccination, more CD8⁺/CD69⁺ cells were observed in both the spleen and tumors in the vaccine group than in the PBS group (p < 0.01, Figures 5E, F ). Interestingly, CD8⁺/CD69⁺ cells showed an obvious reduction in tumor tissue at day 14 compared with day 7 in both groups (p < 0.001, Figure 5F ).

We sought to determine whether the faster and more potent immune response elicited under the fast immunization strategy would result in enhanced antitumor immunity. Further, due to the time-consuming immune response, the normal vaccination strategy has to initiate the treatment 1 day after tumor incubation in our previous research (29, 30, 32). In this part, the fast DNA vaccination strategy was challenged with initiating treatment till tumor mass was about 50-100 mm³. The first DNA vaccine was vaccinated 7 days later in the fast vaccination strategy group than that in the normal vaccination strategy group. The antitumor effect and survival time were analyzed ( Figure 6 ). C57BL/6 mice in the normal vaccination group received the DNA vaccine on day 1 after tumor cell inoculation. Mice in the fast vaccination group received the DNA vaccine on day 8, when tumor volume reached approximately 50 mm³ ( Figure 6A ). Tumor size was suppressed by 47.82% in the fast vaccination group relative to that in the normal immunization group (52.68%). Even though the vaccine was administered 7 days later in the fast immunization group than in the normal immunization group, no significant difference in tumor growth suppression was observed between the two groups (p > 0.05) ( Figure 6B ). Moreover, the survival time of mice was prolonged by 47.45% in the fast immunization group (p < 0.001, vs. control group; p < 0.05 vs. normal group), and the survival time of mice was prolonged by 27.06% in the normal group (p < 0.001, vs. control group, Figure 6C ). These data demonstrate that the fast immunization strategy induced stronger antitumor effects than the normal strategy did, even when vaccines were administered at a later stage.

The safety evaluations tests of the fast DNA vaccination strategy was entrusted to Shandong Xinbo Pharmaceutical R&D, LED. Long-term toxicity test and acute toxicity tests were carried out on Balb/c mice and cynomolgus monkeys. These results showed that no obvious abnormality in weight gain was found in the Balb/c mice after treatment via rapid DNA vaccination strategy, and the major organs (such as heart, liver, kidney, lung, and spleen) of Balb/c mice and cynomolgus monkey showed no toxic effects (data not shown).