Heat shock protein 70 / MAGE-1 tumor vaccine can enhance the potency of MAGE-1-specific cellular immune responses in vivo

Abstract

The cancer-testis antigen encoded by the MAGE-1 gene is an attractive antigen in tumor immunotherapy because it can be processed as a foreign antigen by the immune system and generate tumor-specific cellular immune response in vivo. However, increase of the potency of MAGE-1 DNA vaccines is still needed. The high degree of sequence homology and intrinsic immunogenicity of heat shock protein 70 (HSP70) have prompted the suggestion that HSP70 might have immunotherapeutic potential, as HSP70 purified from malignant and virally infected cells can transfer and deliver antigenic peptides to antigen-presenting cells to elicit peptide-specific immunity. In this research, we evaluated the enhancement of linkage of Mycobacterium tuberculosis HSP70 to MAGE-1 gene of the potency of antigen-specific immunity elicited by naked DNA vaccines. We found that vaccines containing MAGE-1-HSP70 fusion genes enhanced the frequency of MAGE-1-specific cytotoxic T cells in contract to vaccines containing the MAGE-1 gene alone. More importantly, the fusion converted a less effective DNA vaccine into one with significant potency against established MAGE-1-expressing tumors. These results indicate that linkage of HSP70 to MAGE-1 gene may greatly enhance the potency of DNA vaccines, and generate specific antitumor immunity against MAGE-1-expressing tumors.

Fig. 1

Plasmid DNA constructions. pcDNA3.1(+) expression vector with the CMV promoter were used as the basic plasmids. Plasmids pcDNA3-MAGE-1 and pcDNA3-HSP70 were constructed by subcloning the MAGE-1 and HSP70 genes into the downstream of CMV promoters, respectively. The MAGE-1 genes were inserted into the 5′ of M. tuberculosis HSP70 gene of pcDNA3-HSP70 plasmid to construct HSP70 and MAGE-1 fusion DNA tumor vaccine

Fig. 2

ELISpot assays of MAGE-1–specific T-cell precursors from the splenocytes of vaccinated mice restimulated with irradiated B16-MAGE-1. C57BL/6 mice were intramuscularly vaccinated with empty plasmid (pcDNA3), pcDNA3-MAGE-1 DNA (MAGE-1), pcDNA3-HSP70 DNA (HSP), pcDNA3-MAGE-1-HSP70 fusion DNA (MAGE-1-HSP), or pcDNA3-MAGE-1 DNA mixed with pcDNA3-HSP70 DNA (MAGE-1 + HSP). The control group received only PBS. Vaccinations were performed twice with 100 μg of the above gene construction. The number of INF-γ–producing MAGE-1–specific T-cell precursors was determined by using the ELISpot assay. The spot numbers were the mean ± SE in each group. Statistical analysis by a paired Student’s t-test revealed that mice vaccinated with MAGE-1-HSP70 fusion DNA generated the higher IFN-γ⁺ spot number than MAGE-1 DNA (p<0.05, n=6), or a mixture of MAGE-1 DNA and HSP70 DNA (p<0.05, n=6)

Fig. 3

IL-2 secretions of splenocytes restimulated with irradiated B16-MAGE-1 were measured using ELISA. C57BL/6 mice were vaccinated as described in Fig. 1. The splenocytes were pooled 2 weeks after the booster. Splenocytes (4×10⁶) were cocultured with 5×10⁵ irradiated B16-MAGE-1 cells for 3 days in a final volume of 2 ml, and the supernatants were obtained and the concentrations of IL-2 were determined using murine IL-2 ELISA kits. Statistical analysis by a paired Student’s t-test demonstrated that the IL-2 concentration in supernatants of splenocytes from mice vaccinated with MAGE-1-HSP70 fusion DNA was higher than that from mice with MAGE-1 DNA (p<0.05, n=6), or with a mixture of MAGE-1 DNA and HSP70 DNA (p<0.05, n=6)

Fig. 4

MAGE-1–specific lysis against B16-MAGE-1 cells by CTLs induced by vaccination with various DNA. Mice were vaccinated as described in the Fig. 1. The splenocytes of mice vaccinated with various DNA were harvested and restimulated with irradiated B16-MAGE-1 cells for 3 days. The percentage of specific lysis of splenocytes on B16-MAGE-1 target cells was determined by a nonradioactive cytotoxicity assay. Percentage of MAGE-1–specific lysis was calculated by subtracting the percentage lysis of splenocytes on B16 from that on B16-MAGE-1 target cells. The MAGE-1–specific lysis of splenocytes from mice vaccinated with MAGE-1-HSP70 fusion DNA showed a higher level than the other group

Fig. 5

MAGE-1–specifc antibody titers in C57BL/6 mice vaccinated with various recombinant DNA vaccines. C57BL/6 mice were vaccinated as described in Fig. 1. Serum samples were obtained from vaccinated mice 2 weeks after the boost vaccination. The presence of MAGE-1–specific antibody was examined by ELISA. The anti-MAGE-1 sera from mice vaccinated with MAGE-1 protein were used as positive controls. The results of the 1:50 dilution are presented showing the mean absorbance (A _450nm)±SE. No anti-MAGE-1 antibodies could be detected in the sera of mice of any vaccinated group

Fig. 6a–d

The immunotherapy of preestablished B16-MAGE-1 melanoma (a and c) and B16 melanoma (b and d) with DNA vaccination. Groups of mice were s.c. inoculated with a lethal dose of B16-MAGE-1 and B16 tumor cells on day 0 (1×10⁶ cells/mouse, respectively). Mice were immunized with 100 μg of various DNA vaccines (or PBS) on days 3 and 10. Tumor growth kinetics (a and b) were followed over time by caliper measurements, and mean tumor volumes (in mm³) were calculated. Error bars depict SE (n=6 mice/group). For a, MAGE-1-HSP70 fusion DNA vaccines give in statistically significantly better results than all other vaccines from day 24 (p<0.05, n=6), and MAGE-1 DNA and a mixture of MAGE-1 DNA and HSP70 DNA were significantly better than PBS on day 28 (p<0.05, n=6). For b, the difference between MAGE-1-HSP70 versus B16 and MAGE-1-HSP70 versus B16-MAGE-1 was significant from day 21 (p<0.05, n=6). Kaplan-Meier curves (c and d) were generated from survival data (n=6 mice/group). For c, MAGE-1-HSP70 vaccination versus PBS mock vaccination (p<0.01); versus MAGE-1 (p<0.05); versus HSP70 (p<0.05); versus MAGE-1 mixed with HSP70 (p<0.05; while all other comparisons were n.s. Median survival times for each treatment group: PBS, 36 days; pcDNA3, 38 days; MAGE-1, 47 days; HSP70, 41 days; MAGE-1-HSP70, 60 days; MAGE-1 mixed with HSP70, 44 days. For d, MAGE-1-HSP70 vaccination against B16-MAGE-1 tumor cells was significantly better than against B16 (p<0.01). Median survival times were 40 days for MAGE-1-HSP70 against B16, and 36 days for PBS against B16

Fig. 7a, b

Vaccination with MAGE-1-HSP70 DNA protects mice against the growth of MAGE-1–expressing B16 tumor cells. C57BL/6 mice were immunized with empty plasmid pcDNA3, MAGE-1 DNA, HSP70 DNA, MAGE-1-HSP70 DNA and MAGE-1 mixed with HSP70 DNA, or saline (PBS), 21 and 7 days before challenge with melanoma cells. Mice were inoculated with 2×10⁵ B16-MAGE-1 or B16 melanoma cells per mouse subcutaneously in the right leg. The mice were monitored for evidence of tumor growth by palpation and inspection twice a week. a B16-MAGE-1 melanoma cells were s.c. inoculated. All mice vaccinated with pcDNA3, HSP70, or PBS grew tumors within 21 days. Four out of five mice vaccinated with MAGE-1 or MAGE-1 mixed with HSP70 grew tumors. In contrast, all of mice vaccinated with MAGE-1-HSP70 fusion DNA remained tumor-free 50 days after the tumor challenged. b Mice vaccinated with PBS or MAGE-1-HSP70 fusion DNA were challenged with B16 melanoma cells on day 0, and four of five mice treated with MAGE-1-HSP70 fusion DNA and all five mice treated with PBS grew tumors within 17 days