DNA vaccination with all-trans retinoic acid treatment induces long-term survival and elicits specific immune responses requiring CD4⁺ and CD8⁺ T-cell activation in an acute promyelocytic leukemia mouse model

Despite improved molecular characterization of malignancies and development of targeted therapies,¹  acute leukemia is not curable and few patients survive more than 10 years after diagnosis.^2,3  To further improve outcome, we studied the potential efficacy of boosting the patient's immune response. Therapeutic vaccination aimed at promoting T-cell immunity requires: activation of the innate system, optimal presentation of major histocompatibility complex (MHC) class I–binding peptides, and provision of CD4⁺ T-cell help. DNA vaccines have the potential to supply all of these. In an animal model of acute promyelocytic leukemia (APL),⁴  we developed a promyelocytic leukemia-RARα (PML/RARα)–targeted DNA-based vaccine⁵  and show that DNA combined with all-trans retinoic acid (ATRA) has a pronounced survival advantage, concomitant with time-dependent antibody production,^5,6  and an increase in interferon-γ (IFN-γ).⁵  A similar approach confirmed these findings.⁷  The role of ATRA as an immunomodulator is well documented.^8-11  Furthermore, Westervelt et al¹²  show that ATRA responses are influenced by the presence of an intact adaptive immune response.

The present study is aimed at investigating the immune responses involved in the antileukemic effect of the combined ATRA + DNA therapy, particularly those mediated by CD4⁺ and CD8⁺ T cells.

We extended our previous study using the same protocol illustrated in supplemental Figure 1A (available on the Blood website; see the Supplemental Materials link at the top of the online article) where we combined ATRA (5 mg) and a PML-RARαFrC DNA construct in an APL mouse model.⁵  Peripheral blood (PB) was collected approximately every 20 days from day 19 after APL engraftment to follow the mice clinically. At specific days, mice were killed to evaluate responses. Methods are detailed in the figure legends. Animal studies were undertaken according to the guidelines of the institutional animal care committee of Hôpital Saint-Louis.

Our finding of ATRA + DNA treatment significantly extending survival compared with ATRA alone was confirmed in additional protocols (Figure 1A-B; supplemental Figure 1B-C). We have previously reported that DNA alone elicited a modest survival advantage⁵  and that FrC alone failed to give long-term survival¹³  (supplemental Figure 1D).

ATRA + DNA–treated mice had increased white blood cell counts on day 21 after leukemia engraftment, possibly resulting from the increases in CD4⁺ and CD8⁺ cells (supplemental Figure 2A). ATRA- and nearly half of the ATRA + DNA–treated mice had reduced platelets by day 60 (supplemental Figure 2A). Long-term survivors (LTSs) from the ATRA + DNA–treated group (> 120 days) had significantly reduced bone marrow blast counts (2%-12%; supplemental Figure 2B). ATRA-alone–treated mice died by day 90.

PB absolute counts of CD3⁺CD4⁺ and CD3⁺CD8⁺ subsets, as well as natural killer (NK), Tγδ, and B lymphocytes, were measured by flow cytometry. Mice treated with ATRA alone had lower CD3⁺CD4⁺ and CD3⁺CD8⁺ absolute counts compared with ATRA + DNA–treated mice at day 21 (supplemental Figure 3). The increased CD3⁺CD4⁺ and CD3⁺CD8⁺ absolute counts in ATRA + DNA–treated animals were associated with improved survival (supplemental Figure 3A-B), probably reflecting mobilization of immune responses.

Increasing the number of T cells of ATRA-treated mice by infusion of CD3⁺CD4⁺ cells originated from ATRA- or ATRA + DNA–treated mice did not rescue these mice from relapse and death (supplemental Figure 4). This indicates that the protective effect of the combined therapy is not quantitative but qualitative.

No differences of NK, Tγδ, and B cells in the PB were detected among untreated, ATRA-, or ATRA + DNA–treated APL mice (data not shown).

To examine the implication of CD4⁺ or CD8⁺ T cells, leukemic mice were depleted of either CD4 or CD8⁺ T cells.

Injections of CD4⁺ or CD8⁺ subset-specific monoclonal antibody (mAb) starting after the first DNA vaccination did not affect the survival of untreated (data not shown), ATRA-treated, or rat IgG–treated vaccinated mice (isotype control; Figure 1A-B). The depletion of CD4⁺ or CD8⁺ subsets completely abolished the antileukemic effect of the combined therapy. None of the CD4- or CD8-depleted ATRA + DNA–treated mice survived after day 100 (Figure 1A, P < .001; and Figure 1B, P = .045, respectively). These data reflect the need of these T cells for the generation of DNA vaccine driven anti–APL-protective immune responses.

Depletions of CD4⁺ T cells in LTSs from the ATRA + DNA–treated group led to relapse and death of all mice (n = 6) between 26 and 83 days after the initiation of depletion (Figure 1C). All CD8-depleted LTSs as well as those treated with rat IgG were alive on day 160 after depletion. These results show the crucial role of CD4⁺ T cells in the maintenance of the antileukemic effect of DNA vaccination and suggest that, although these mice are in complete remission, they still need CD4⁺ T-cell activity for maintenance.

To evaluate the cytotoxic activity, we used the carboxyfluorescein diacetate succinimidyl ester (CFSE)–based assay.¹⁴  Effectors from ATRA-treated (Figure 2A) or ATRA + DNA–treated (Figure 2B) APL mice had specific cytotoxic activity against APL assayed on days 34 to 57. Effectors from LTSs from ATRA + DNA–treated mice showed less than 60% target survival at a ratio of 50:1 (Figure 2C). To demonstrate that the killing of APL cells was MHC-restricted, we performed cytotoxic assays in the presence of blocking mAb. The specific cytotoxicity observed with the isotype control Ab was significantly reduced when APL target cells were preincubated with the anti–H-2D^q/L^q Ab (increased survival from 52% with isotype control Ab to 80% with H-2^q–specific mAb at a 25:1 effector/target [E:T] ratio), thus demonstrating that the cytotoxic activity was MHC-restricted (Figure 2D). The remaining cytotoxic activity observed in the presence of the blocking mAb could reflect the residual H-2K^q MHC-restricted activity or nonrestricted cytotoxicity mediated by NK or Tγδ cells.

Because degranulation of cytotoxic T cells and killing of targets are correlated,^15,16  we performed the CD107a mobilization assay using LTS splenocytes. APL-specific activation of CD3⁺CD8⁺ T cells (1.5%-17%) was observed (supplemental Figure 5E).

CD3⁺CD8⁺CD107a⁺ sorted cells from an ATRA + DNA–treated LTSs induced cytotoxicity, with less than 25% of viable APL targets rescued at a 1:1 E/T ratio (Figure 2E) demonstrating APL-specific cytotoxic activity.

DNA vaccines usually stimulate T_H1 cytokines. The cytokine release profile of ATRA + DNA–treated LTSs was analyzed earlier than 300 days assayed previously.⁵  Analysis showed APL-specific increases in T_H1 cytokines (23-fold for IFN-γ and 5-fold for TNF-α; supplemental Figure 6).

To assess priming of APL-specific T cells, effectors from ATRA- and ATRA + DNA–treated APL mice were assayed in an ELISpot IFN-γ assay. None of the ATRA-treated APL mice showed any substantial secretion of IFN-γ secretion compared with the baseline secretion. In contrast, DNA vaccine induced IFN-γ secretion (Figure 2F). Interestingly, when the LTSs from the ATRA + DNA group were assayed, increases of IFN-γ–producing cells were observed (supplemental Figure 7).

Overall, using complementary methods, we show that APL-specific T cells remained active in vaccinated animals long after the last boost of DNA. The aim of DNA vaccination is to target tumor cells not eradicated by current protocol, preferably in the setting of minimal disease load. The power of the immune system is clear from the effectiveness of passive immunity. DNA vaccination as an approach for immunotherapy is increasing.^17,18  Although ATRA may be a good adjuvant for boosting immune responses,^10,11,19,20  the mounted immune responses were insufficient to contain the disease. Combining DNA with ATRA resulted in increased APL-specific immune responses, with MHC-restricted CD8⁺ T-cell responses and increased IFN-γ production, which rescued the disease. Together with the finding of a major role for CD4⁺ T cells in maintaining the durable remissions, these data suggest that, in human clinical trials, the combination of ATRA + DNA vaccine should control minimal residual disease. This study provides insights into the immunology of DNA vaccines, and this may be relevant for targeting other malignancies.

The authors thank Scott Kogan and Michael Bishop (University of California–San Francisco) for the PML-RARα transgenic mouse cells; Freda Stevenson for the FrC construct; Martine Chopin and Aurore Cleret for technical assistance; and members of the Institut Universitaire d'Hématologie—specifically Michel Schmid from the Imagery Department—for flow sorting, Bernard Boursin from the photography laboratory, and members of the Département d'Expérimentation Animale for animal care.

This work was supported by Cancéropôle axe 3 Ile de France, Association pour la Recherche sur le Cancer no. 3820, Fondation de France, Cent pour Sang la Vie, the European Leukemia Network, and Inserm. K.F. was a postdoctoral fellow funded by the Sumimoto Life Social Welfare Services and Foundation and Fondation de France. K.P. was a graduate student funded by a grant (no. 94308) from Grant Agency of Charles University, Prague, Czech Republic and by a French Government Grant. C.L.P. was a graduate student funded by Cancéropôle. M.A. was a postdoctoral fellow funded by Fondation de France. R.A.P. is a member of the European COST action BM0801.

Contribution: K.F. and K.P. performed research and analyzed data; C.L.P., M.R., V.B., and P.K. performed research; M.A. performed research and contributed to statistical design and analysis; A.J. reviewed the histology slides; M.-E.N. reviewed the bone marrow slides; R.W. contributed to statistical design and analysis; D.C. and C.C. designed research and wrote the paper; and M.P., H.M.-T., and R.A.P. designed and performed research and wrote the paper.

Conflict-of-interest disclosure: R.A.P., C.C., and D.C. have patents pending through Inserm related to technology employed in this present study. R.A.P. and C.C. are founding members of and have financial interest in a company (Vivavacs) and are members of the executive and scientific boards negotiating rights to these same patents. The remaining authors declare no competing financial interests.

Correspondence: Rose Ann Padua, Inserm UMRS 940, Institut Universitaire d'Hématologie, Hopital Saint-Louis, 1 Ave Claude Vellefaux, 75010 Paris, France; e-mail: rose-ann.padua@inserm.fr.