Novel Exosome-Targeted CD4⁺ T Cell Vaccine Counteracting CD4⁺ 25⁺ Regulatory T Cell-Mediated Immune Suppression and Stimulating Efficient Central Memory CD8⁺ CTL Responses

Abstract

T cell-to-T cell Ag presentation is increasingly attracting attention. In this study, we demonstrated that active CD4⁺ T (aT) cells with uptake of OVA-pulsed dendritic cell-derived exosome (EXO_OVA) express exosomal peptide/MHC class I and costimulatory molecules. These EXO_OVA-uptaken (targeted) CD4⁺ aT cells can stimulate CD8⁺ T cell proliferation and differentiation into central memory CD8⁺ CTLs and induce more efficient in vivo antitumor immunity and long-term CD8⁺ T cell memory responses than OVA-pulsed dendritic cells. They can also counteract CD4⁺25⁺ regulatory T cell-mediated suppression of in vitro CD8⁺ T cell proliferation and in vivo CD8⁺ CTL responses and antitumor immunity. We further elucidate that the EXO_OVA-uptaken (targeted)CD4⁺ aT cell’s stimulatory effect is mediated via its IL-2 secretion and acquired exosomal CD80 costimulation and is specifically delivered to CD8⁺ T cells in vivo via acquired exosomal peptide/MHC class I complexes. Therefore, EXO-targeted active CD4⁺ T cell vaccine may represent a novel and highly effective vaccine strategy for inducing immune responses against not only tumors, but also other infectious diseases.

Dendritic cells (DC)³ process exogenous Ags in endosomal compartments such as multivesicular endosomes (1) which can fuse with plasma membrane, thereby releasing Ag-presenting vesicles called “exosomes” (EXO) (2, 3). EXO are 50- to 90-nm diameter vesicles containing Ag-presenting (MHC class I, class II, CD1, heat shock protein 70–90), tetraspan (CD9, CD63, CD81), adhesion (CD11b, CD54), and costimulatory (CD86) molecules (4, 5), i.e., the necessary machinery required for generating potent immune responses.

Zitvogel et al. (3) first demonstrated vaccination of DC-derived EXO in eradication of tumors in animal models. Subsequently, EXO-based vaccines have been confirmed to stimulate strong CTL responses and induce antitumor immunity in different animal models (6, 7). However, its therapeutic efficiency is still limited to only production of prophylactic immunity against tumors possibly due to lacking capacity of breaking immune tolerance (8). Exosomal peptide/MHC class I and II (pMHC I and II) complexes are functional, but require transfer to DC to promote T cell activation leading to tumor eradication (9–11). Therefore, the potential pathway of in vivo EXO-mediated antitumor immunity may be through uptake of EXO by host immature DC that, in turn, stimulate Ag-specific T lymphocytes via the pMHC complexes and costimulatory molecules on EXO-uptaken DC. In addition, Kennedy et al. (12) have also demonstrated that CD4⁺ T cells can acquire APC membrane molecules in vivo and induce memory CTL responses. We have recently demonstrated that mature DC with uptake of OVA-pulsed DC (DC_OVA)-derived EXO (EXO_OVA) express a higher level of pMHC I and costimulatory CD40, CD54, and CD80 molecules and can strongly stimulate OVA-specific CD8⁺ CTL responses and antitumor immunity (13). Recently, we have also demonstrated that CD4⁺ Th cells can acquire membrane molecules from DC via DC activation, and act as Th-APCs (14). These Th-APC with acquired pMHC I and costimulatory CD54 and CD80 molecules can stimulate tumor-specific CD8⁺ CTL responses and induce antitumor immunity. One of the potential mechanisms of CD4⁺ T cell acquisition of DC molecules is uptake of DC-released EXO by CD4⁺ T cells. Therefore, these results clearly indicate that DC-derived EXO can transfer the Ag-presenting activity of DCs to either DC or CD4⁺ T cells through EXO uptake by these cells. However, the EXO-targeted CD4⁺ T cell vaccine and its potential immune mechanism in induction of antitumor CTL responses have not been studied and elucidated.

In this study, we demonstrated that EXO_OVA-uptaken (targeted) active CD4⁺ T (aT_EXO) cells can 1) stimulate central memory CD8⁺ CTL responses, more efficient antitumor immunity, and T cell memory than DC_OVA and 2) counteract CD4⁺25⁺ regulatory T (Tr) cell-mediated immune suppression. We also elucidated that the aT_EXO stimulatory effect is mediated via its IL-2 secretion and acquired exosomal CD80 costimulation and is specifically delivered to CD8⁺ T cells in vivo via acquired exosomal pMHC I complexes.

Materials and Methods

Reagents, cell lines, and animals

OVA was obtained from Sigma-Aldrich. OVA I (SIINFEKL) and OVA II (ISQAVHAAHAEINEAGR) which are OVA peptides specific for H-2K^b and Ia^b, respectively (15, 16). Mut I (FEQNTAQP) peptide is specific for H-2K^b of an irrelevant 3LL lung carcinoma (11). All peptides were synthesized by Multiple Peptide Systems. Biotin-labeled or FITC-labeled Abs specific for H-2K^b (AF6-88.5), Ia^b (AF6-120.1), CD4 (GK1.5), CD8 (53-6.7), CD11c (HL3), CD25 (7D4), CD40 (IC10), CD44 (IM7), CD54 (3E2), CD62L (MEL-14), CD69 (H1.2F3), CD80 (16-10A1), IL-7R (4G3), and Vα2Vβ5⁺ TCR (MR9-4), as well as FITC-conjugated avidin, were all obtained from BD Pharmingen. Anti-LFA-1 Ab, CTLA-4/Ig fusion protein, recombinant mouse IL-4, and GM-CSF were purchased from R&D Systems. The mouse thymoma cell line EL4 and OVA-transfected EL4 (EG7) cell line were obtained from American Type Culture Collection. The highly lung metastasis OVA-transfected BL6 –10_OVA melanoma cell line was generated in our own laboratory (14). Female C57BL/6 (B6, CD45.2⁺), C57BL/6.1 (B6.1, CD45.1⁺), OVA-specific TCR-transgenic OT I and OT II mice, H-2K^b, Ia^b, IL-2, IFN-γ, TNF-α, CD40, CD54, and CD80 gene knockout (KO) mice were obtained from The Jackson Laboratory. Homozygous OT II/H-2K^b−/−, OT II/CD40^−/−, OT II/CD54^−/−, OT II/CD80^−/−, OT II/IL-2^−/−, OT II/IFN-γ, and OT II/TNF-α ^−/− mice were generated by backcrossing the designated gene KO mice onto the OT II background. All mice were treated according to animal care committee guidelines of the University of Saskatchewan.

DC generation

DCs were generated as described previously (17). Briefly, spleen cells were prepared in PBS with 5 mM EDTA, washed, and incubated in culture medium with 7% FCS at 37°C for 2 h. Nonadherent cells were removed by gentle pipetting with warm serum-free medium. Adherent cells were cultured overnight in medium with 1% normal mouse serum, GM-CSF (1 ng/ml), and OVA (0.3 mg/ml). These DCs were termed as DC_OVA. DC generated from H-2K^b, CD54, CD40, and CD80 gene KO mice were referred to as (K^b−/−)DC_OVA, (CD40^−/−)DC_OVA, (CD54^−/−)DC_OVA and (CD80^−/−)DC_OVA, respectively.

EXO preparation

Preparation and purification of EXO derived from the culture supernatants of DC_OVA were previously described (13), an overage of 5 μg of EXO was recovered from an overnight culture of 1 × 10⁶ DC_OVA in 1.0 ml of AIM-V serum-free medium containing GM-CSF (10 ng/ml). EXO derived from DC_OVA were termed EXO_OVA. Similar to DC_OVA, EXO_OVA also expressed MHC class I (H-2K^b) and class II (Ia^b), CD11c, CD40, CD54, CD80, and the pMHC I complex, but in a lesser content, compared with DC_OVA (13). EXO derived from (K^b−/−) DC_OVA, (CD40^−/−)DC_OVA, (CD54^−/−) DC_OVA, and (CD80^−/−)DC_OVA were termed (K _b^−/−)EXO_OVA, (CD40^−/−)EXO_OVA, (CD54^−/−)EXO_OVA, and (CD80^−/−)EXO_OVA, respectively. To generate CFSE-labeled EXO, DC_OVA were stained with 0.5 μM CFSE at 37°C for 20 min (16) and washed three times with PBS, and then pulsed with OVA protein in AIM-V serum-free medium overnight. The CFSE-labeled EXO (EXO_CFSE) was harvested and purified from the culture supernatants as previously described above.

CD4⁺ T cell preparation and characterization

Naive OVA-specific T (nT) cells were isolated from OVA-specific TCR-transgenic OT I and OT II mouse spleens, enriched by passage through nylon wool columns (C&A Scientific), and then purified by negative selection using anti-mouse CD8(Ly2) or CD4 (L3T4) paramagnetic beads (Dynal Biotech) to yield populations that were >98% CD4⁺/Vα2Vβ5⁺ or CD8⁺/Vα2Vβ5⁺, respectively (14). To generate active OT II CD4⁺ T cells, the spleen cells from OT II mice were cultured in RPMI 1640 medium containing IL-2 (20 U/ml) and Con A (1 μg/ml) for 3 days. The Con A-activated CD4⁺ T (aT) cells were purified using nylon wool columns and then CD4 microbeads (Miltenyi Biotec). The CD4⁺ aT cells derived from OT II/H-2K^b−/−, OT II/CD40^−/−, OT II/CD54^−/−, and OT II/ CD80^−/−mice were termed aT(H-2K^b−/−), aT(CD40^−/−), aT(CD54^−/−), and aT(CD80^−/−), respectively. The above naive and Con A-activated CD4⁺ T cells were analyzed using a panel of Abs by flow cytometry. To assess the potential counteraction of CD4⁺25⁺ Tr cell-induced suppression by EXO-targeted CD4⁺ T cells, CD4⁺CD25⁺ Tr cells were purified from C57BL/6 mouse spleen T cells using CD25 microbeads (clone 7D4; Milte-nyi Biotec) (18). The purified CD4⁺CD25⁺ Tr cells were stained with PE-anti-CD25 (clone PC61.5), FITC-CD4 (clone GK1.5), and energy-coupled dye (ECD)-Foxp3 (clone FJK-16S) using Regulatory T Cell Staining kit No. 3 (eBioscience), and then analyzed by flow cytometry.

Exosomal molecule uptake by CD4⁺ T cells

First, the OT II CD4⁺ nT and aT cells were incubated with EXO_CFSE (10 μg/1 × 10⁶ T cells in 200 μl of AIM-V serum-free medium containing IL-2 (20 U/ml)) at 37°C for 5 h and then analyzed for CFSE staining by flow cytometry. In another set of experiments, the OT II CD4⁺ aT cells were incubated with EXO_CFSE under the above same condition for 1, 3, and 5 h, and then assessed by confocal fluorescence microscopy. To further determine the transfer of exosomal molecules to T cells, the CD4⁺ nT and aT cells from OT II mice or OT II mice with different gene KO were incubated with EXO_OVA, and then analyzed for expression of H-2K^b, CD40, CD54, CD80, and pMHC I by flow cy-tometry or analyzed for expression of CD40 and CD80 by confocal fluorescence microscopy. The OT II CD4⁺ nT and aT cells cocultured with EXO_OVA were termed nT_EXO and aT_EXO cells, respectively. The CD4⁺ aT(H-2K^b−/−), aT(CD40^−/−), aT(CD54^−/−), and aT(CD80^−/−) cells were cocultured with (K^b−/−)EXO_OVA, (CD40^−/−)EXO_OVA, (CD54^−/−)EXO_OVA, and (CD80^−/−)EXO_OVA and termed CD4⁺ aT_EXO (K^b−/−), aT_EXO(CD40^−/−), aT_EXO(CD54^−/−), and aT_EXO(CD80^−/−) cells, respectively. The CD4⁺ aT cells from OT II/IL-2^−/−, OT II/ IFN-γ, and OT II/TNF-α ^−/− mice were cocultured with EXO_OVA and termed CD4⁺ aT_EXO(IL-2^−/−), aT_EXO(IFN-γ ^−/−), and aT_EXO(TNF-α ^−/−) cells, respectively. The cytokine profiles of aT_EXO(K ^b−/−), aT_EXO (CD40^−/−), aT_EXO(CD54^−/−), and aT_EXO(CD80^−/−) cells are similar to that of aT_EXO cells, whereas the cytokine profiles of aT_EXO(IL-2^−/−), aT_EXO(IFN-γ ^−/−), and aT_EXO(TNF-α ^−/−) cells are also similar to that of aT_EXO cells except for the specific cytokine (IL-2 or IFN-γ or TNF-α) deficiency (data not shown). For blocking assays, OT II CD4⁺ aT cells were incubated with anti-H-2K^b, anti-Ia^b, and anti-LFA-1 Abs (12 μg/ml) or CTLA-4/Ig (12 μg/ml), respectively, on ice for 30 min, then were cocultured with EXO_CFSE for 5 h at 37°C. The cells were harvested and analyzed for CFSE expression by flow cytometry.

T cell proliferation assays

To assess the functional effect of CD4⁺ nT_EXO and aT_EXO cells, we performed an in vitro CD8⁺ T cell proliferation assay. CD4⁺ nT_EXO and aT_EXO (0.3 × 10⁵ cells/well) cells derived from OT II mice and their 2-fold dilutions were cultured with a constant number of naive OT I CD8⁺ T cells (1 × 10⁵ cells/well) in presence or absence of CD4⁺CD25⁺ Tr cells (0.3 × 10⁵ cells/well) derived from C57BL/6 mice. To examine the molecular mechanism, a panel of reagents including anti-H-2K^b, I-A^b, and LFA-1 Abs and CTLA-4/Ig fusion protein (each 10 μg/ml), a mixture of the above reagents (as mixed reagents), and a mixture of isotype-matched irrelevant Abs (as control reagents) were added to the cell cultures, respectively. After culturing for 3 days, thymidine incorporation was determined by liquid scintillation counting (19). In the in vivo OVA-specific CD8⁺ T cell proliferation assay, C57BL/6 mice (six mice per group) were i.v. immunized with irradiated (4000 rad) stimulators including DC_OVA, CD4⁺ aT_EXO cells (1 × 10⁶ cells/mouse), or CD4⁺ aT_EXO cells (1 × 10⁶ cells/ mouse) with different gene KO. In another set of experiments, C57BL/6 mice (six mice per group) were i.v. immunized with irradiated (4000 rad) DC_OVA, CD4⁺ aT, and aT_EXO cells (1 × 10⁶ cells/mouse) alone or together with CD4⁺25⁺ Tr cells (3 × 10⁶ cells/mouse). For evaluation of in vivo OVA-specific CD8⁺ T cell proliferation, the tail blood samples derived from mice 6 days after immunization were incubated with PE-H-2K^b/ OVAI tetramer and FITC-anti-CD8 Ab (Beckman Coulter) and analyzed by flow cytometry (13).

Cytotoxicity assay

In vivo cytotoxicity assay was performed as previously described (14). Briefly, wild-type C57BL/6 or Ia^b−/− mice (six mice per group) were i.v. immunized with the above stimulator cells (1 × 10⁶ cells/mouse), respectively. In another set of experiments, C57BL/6 mice (six mice per group) were i.v. injected with irradiated (4000 rad) DC_OVA and CD4⁺ aT_EXO cells (1 × 10⁶ cells/mouse) alone or together with CD4⁺25⁺ Tr cells (3 × 10⁶ cells/mouse). Splenocytes were harvested from naive mouse spleens and incubated with either high (3.0 μM, CFSE^high) or low (0.6 μM, CFSE^low) concentrations of CFSE, to generate differentially labeled target cells. The CFSE^high cells were pulsed with OVA I peptide, whereas the CFSE^low cells were pulsed with Mut 1 peptide and served as internal controls. These peptide-pulsed target cells were washed extensively to remove free peptides, and then i.v. coinjected at a 1:1 ratio into the immunized mice 6 days after immunization. Sixteen hours after the target cell delivery, the spleens of immunized mice were removed and residual CFSE^high and CFSE^low target cells remaining in the recipients’ spleens were analyzed by flow cytometry.

Animal studies

To examine the antitumor protective immunity conferred by EXO-targeted CD4⁺ T cells, wild-type C57BL/6 mice (n = 8) were injected i.v. with irradiated (4000 rad) DC_OVA, aT, and aT_EXO cells (1 × 10⁶ cells/mouse) with or without coinjection of CD4⁺25⁺ Tr cells (3 × 10⁶ cells/mouse) aT_EXO cells with various gene KO (1 × 10⁶ cells/mouse), respectively. The mice were injected with PBS as a control. To assess the antitumor immunity, the immunized mice were challenged i.v. with 0.5 × 10⁶ BL6–10_OVA tumor cells 6 days subsequent to the immunization. To assess the long-term antitumor immunity, wild-type C57BL/6 mice were first immunized with irradiated (4000 rad) DC_OVA and aT_EXO cells (1 × 10⁶ cells/mouse). The immunized mice were then challenged i.v. with 2 × 10⁶ BL6–10_OVA cells 3 mo subsequent to the immunization to assess development of tumor-specific CD8⁺ memory T (Tm) cells. The mice were sacrificed 4 wk after tumor cell injection and two individuals counted the lung metastasis tumor colonies in a blind fashion. BL6–10_OVA-derived metastases on freshly isolated lungs appeared as discrete black-pigmented foci that could be distinguishable from normal lung tissues and all were confirmed by histological examination. Metastasis foci too numerous to count were assigned an arbitrary value of >100 (14).

Results

CD4⁺ T cells uptake exosomal pMHC I and costimulatory molecules in both Ag-specific and nonspecific manners

The nT and Con A-stimulated aT cells derived from transgenic OT II mice expressed both CD4 and TCR. The latter ones also expressed active T cell markers (CD25 and CD69) (Fig. 1) and secreted IL-2 (~2.4 ng/ml/10⁶ cells/24 h), IFN-γ (~2.0 ng/ml/10⁶ cells/24 h), and TNF-α (~1.7 ng/ml/10⁶ cells/24 h), but no IL-4 and IL-10, indicating that they are active type Th1 cells. In addition, there was no CD11-positive splenic DC contamination in the purified CD4⁺ aT population (Fig. 1). To assess EXO uptake, CD4⁺ nT and aT cells derived from OT II and wild-type C57BL/6 (B6) mice were incubated with EXO_CFSE expressing CFSE (Fig. 2a) for various times and then analyzed by flow cytometry and confocal fluorescence microscopy. As shown in Fig. 2a, the CFSE dye was minimally and apparently detectable on OT II CD4⁺ nT and aT cells, respectively. The uptake of EXO_CFSE by OT II CD4⁺ aT cells increased with incubation time and reached a maximal level (80% CFSE-positive cells) after a 5-h incubation (Fig. 2b), which was also confirmed by confocal fluorescence microscopic analysis (Fig. 3a). The CFSE-positive cells declined with time when culturing them in medium, but were still detectable >3 days in culture, indicating that the uptaken exosomal molecules on CD4⁺ T cells are quite stable, which is consistent with a previous report by Undale et al. (20). Similar to CFSE dye, other exosomal molecules such as H-2K^b, CD40, CD54, and CD80 molecules were also minimally and apparently transferred from EXO_OVA onto OT II CD4⁺ nT and aT cells, respectively (Fig. 2, c and e). In addition, the pMHC I complexes were also transferred onto CD4⁺ aT cells. To further confirm it, we incubated EXO_OVA with OT II CD4⁺ T cells with different gene KO and then analyzed by flow cytometry or confocal fluorescence microscopy. The original OT II CD4⁺ nT and aT cells with different gene KO did not express endogenous H-2K^b, CD40, CD54, and CD80, respectively. However, after incubation with EXO_OVA, T cells (especially the CD4⁺ aT cells) did display the above exosomal molecules (Fig. 2, d and f), indicating that increased expression of the above molecules on CD4⁺ T cells is due to an uptake of EXO molecules rather than an endogenous up-regulation. This was also confirmed by confocal fluorescence microscopic analysis. As shown in Fig. 3b, CD4⁺ aT cells with deficiency of endogenous CD40 and CD80 expression displayed CD40 and CD80 expression after uptake of EXO_OVA. To elucidate the molecular mechanism involved in EXO uptake, we used a panel of reagents in a blocking assay. The anti-Ia^b and LFA-1 Abs, but not the CTLA-4/Ig fusion protein and the anti-H-2K^b Ab, were able to significantly block EXO uptake ( p < 0.05) (Fig. 2g). This was further confirmed by a significant decrease in EXO uptake ( p < 0.05) when incubation of aT(CD54^−/−), but not aT(CD40^−/−) and aT(CD80^−/−), cells with EXO_CFSE derived from CFSE-labeled (CD54^−/−)DC_OVA, (CD40^−/−)DC_OVA, and (CD80^−/−) DC_OVA, respectively, indicating that the EXO_OVA uptake by CD4⁺ T cells is mediated by both OVA-specific Ia^b/TCR and nonspecific CD54/LFA-1 interactions, but not by H-2K^b/TCR and CD80/CD28 interactions, which is consistent with previous reports by Hwang et al. (21, 22).

FIGURE 1.

Phenotypic analysis of CD4⁺ T cells by flow cytometry. Flow cytometric analysis of OT II CD4⁺ T cells. The nT OT II and aT OT II cells (thick solid lines) were stained with a panel of Abs and then analyzed by flow cytometry. These cells were also stained with isotype-matched irrelevant Abs, respectively, and used as control populations (thin dotted lines). One representative experiment of three is shown.

FIGURE 2.

EXO uptake by CD4⁺ T cells. a, Both naive and active OT II and C57BL/6 CD4⁺ T cells with (thick solid lines) and without (thin dotted lines) uptake of EXO_CFSE were analyzed for CFSE expression by flow cytometry. b, Kinetic study of EXO uptake by CD4⁺ T cells. Active OT II CD4⁺ T cells (2 × 10⁶ cells) were incubated with EXO_CFSE (20 μg) in DMEM (0.2 ml) containing 10% FCS and IL-2 (10 U/ml) at 37°C for different times. CFSE-positive T cells were detected by fluorescence microscopy and the percentages of CFSE-positive T cells were calculated. In another set of experiments, active OT II CD4⁺ T cells with incubation of EXO_CFSE for 5 h were cultured in DMEM containing 10% FCS and IL-2 (10 U/ml) at 37°C for different times. CFSE-positive T cells were detected by fluorescence microscopy and the percentages of CFSE-positive T cells were calculated. c and e, Both naive and active OT II CD4⁺ T cells with (thick solid lines) and without (thick dotted lines) uptake of EXO_OVA were analyzed for expression of a panel of surface molecules including H-2K^b, CD40, CD54, CD80, and pMHC I by flow cytometry. Irrelevant iso-type-matched Abs were used as controls (thin dotted lines). d and f, Both naive and active OT II CD4⁺ cells from H-2K^b, CD40, CD54, and CD80 gene KO mice were also cocultured with (thick solid lines) and without (thick dotted lines) EXO_OVA and then analyzed for expression of H-2K^b, CD40, CD54, and CD80 by flow cytometry, respectively. Irrelevant isotype-matched Abs were used as controls (thin dotted lines). g, In the blocking assay, OT II CD4⁺ aT cells were treated with anti-H-2K^b, anti-Ia^b, and anti-LFA-1 Abs, and CTLA-4/Ig fusion protein, a mixture of these reagents or a mixture of matched isotype Abs (as control reagents) or PBS (as positive control) on ice for 30 min, respectively, and then incubated with EXO_CFSE. In addition, OT II CD4⁺ aT cells with CD54 deficiency were incubated with EXO_CFSE derived from CFSE-labeled (CD54^−/−) DC_OVA. The fractions of CFSE-positive T cells were analyzed after coculture for 5 h at 37°C. *, p < 0.05 vs cohorts of the positive control (aT/EXO_CFSE) (Student’s t test). One representative experiment of three is displayed.

FIGURE 3.

Membrane molecule acquisition analysis by confocal fluorescence microscopy. a, CD4⁺ aT cells derived from OT II mice were incubated with EXO_CFSE for 1, 3, and 5 h and then examined under differential interference contrast (DIC) and by confocal fluorescence microscopy. b, CD4⁺ aT cells derived from OT II mice with CD40 or CD80 gene deficiency were incubated with EXO_OVA for 5 h, stained with FITC-conjugated anti-CD40 or anti-CD80 Ab, and then examined under DIC and by confocal fluorescence microscopy. One representative experiment of two is displayed.

CD4⁺ aT_EXO cells counteract in vitro CD4⁺25⁺ Tr cell-mediated suppression of naive CD8⁺ T cell proliferation

We then examined the stimulatory effect of EXO_OVA-targeted active CD4⁺ T (aT_EXO) cells. As shown in Fig. 4a, EXO_OVA could prime CD8⁺ T cell proliferation in vitro, which is consistent with previous reports by Hwang et al. (21, 22), but in to a lesser extent compared with DC_OVA. Of interest, CD4⁺ aT_EXO is a stronger stimulator in CD8⁺ T cell proliferation than DC_OVA, whereas nT_EXO is a relatively weak stimulator. CD4⁺25⁺ Tr cells expressing CD4, CD25, and Foxp3 (Fig. 4b) inhibited DC_OVA-stimulated CD8⁺ T cell proliferation (Fig. 4a). However, aT_EXO cells maintained their stimulatory effect in presence of CD4⁺25⁺ Tr cells, indicating that aT_EXO cells can counteract in vitro CD4⁺25⁺ Tr cell-mediated suppression of naive CD8⁺ T cell proliferation.

FIGURE 4.

Stimulation of CD8⁺ memory T cell responses in vitro. a, In vitro CD8⁺ cell proliferation assay. EXO_OVA (10 μg/ml), DC_OVA, nT_EXO, aT_EXO, and Con A-activated OTII T (aT) cells and their 2-fold dilutions were cocultured with a constant number of OT I CD8⁺T cells in presence or absence of CD4⁺25⁺ Tr cells. After 3 days, the proliferation response of CD8⁺ T cells was determined by [³H]thymidine uptake assay. b, Phenotypic analysis of CD4⁺25⁺ Tr cells. The CD4⁺25⁺ Tr cells were purified from C57BL/6 mouse splenocytes using CD25 microbeads, stained with PE-CD25, FITC-CD4, and ECD-Foxp3 Abs, and then analyzed by flow cytometry. The FITC-CD4- and PE-CD25-positive T cells were grouped for analysis of ECD-Foxp3 expression (solid line). Irrelevant isotype-matched Ab was used as a control (dotted line). c, Phenotypic analysis of in vitro aT_EXO-primed CD8⁺ T cells. CFSE-labeled naive T cells derived from OT I mice were primed with irradiated DC_OVA and aT_EXO for 2 days in vitro and stained for CD8, CD25, CD44, CD62L, and IL-7R, respectively. Dot plots of CFSE-positive CD8⁺ T cells stained with PE-anti-CD8 Ab are shown, indicating that the CFSE-labeled CD8⁺ T cells underwent some cycles of cell division, and were sorted by flow cytometry for assessment of CD25, CD44, CD62L, and IL-7R expression using PE-labeled Abs (solid lines) or PE-isotype-matched irrelevant Abs (dotted lines) by flow cytometry. d, The in vitro DC_OVA- and aT_EXO-activated OT I CD8⁺CD45.1⁺ T cells were purified using biotin anti-CD45.1 Ab and anti-biotin microbeads (Miltenyi Biotec) and are referred to as DC_OVA/OT I_6.1 and aT_EXO/OT I_6.1, respectively. They were then incubated with irradiated (4,000 rad) EG7 and EL4 for 24 h. The supernatants in wells containing DC_OVA/OT I_6.1 plus EG7 or EL4 cells (DC_OVA/OT I_6.1:EG7 or DC_OVA/OT I_6.1:EL4) and aT_EXO/OT I_6.1 plus EG7 or EL4 cells (aT_EXO/OT I_6.1:EG7 or aT_EXO/OT I_6.1:EL4) were examined for IFN-γ expression by ELISA. e, T cell proliferation assay. In vitro DC_OVA- and aT_EXO-activated CD8⁺CD45.1⁺ T cells (0.4 × 10⁵ cells/well) derived from OT I/B6.1 mouse OTI CD8⁺ T cells, primed on day 0 with irradiated DC_OVA (■) or aT_EXO (▴), were maintained in cultures for 1 wk with the indicated cytokines (IL-2 (50 U/ml) and IL-7 (10 ng/ml)) added on days 3 and 5. Live CD8⁺ T cells with trypan blue exclusion for each culture done in triplicate were counted at the indicated time points. f, In vitro cytotoxicity assay. The above DC_OVA/OT I_6.1 (■) and aT_EXO/OT I_6.1 (▴) cells were used as effector cells, whereas ⁵¹Cr-labeled EG7 or EL4 cells were used as target cells in a chromium release assay. One representative experiment of three is displayed.

CD4⁺ aT_EXO cells stimulate in vitro CD8⁺ T cell proliferation and differentiation into central memory T cells

We next conducted phenotypic characterization of the above in vitro aT_EXO-primed CD8⁺ T cells. Our data showed that DC_OVA or aT_EXO priming resulted in several cycles of CD8⁺ CFSE-T cell division, whereas nontreated CD8⁺ CFSE-T cells were not divided (Fig. 4c). The primed T cells displayed expression of CD25, CD44 (Tm marker) (23) and CD62L (Fig. 4c). However, aT_EXO-primed CD8⁺ T cells expressed IL-7R and higher CD62L than DC_OVA-primed ones with no IL-7R expression, indicating that they may be prone to becoming long-lived Tm cells. We then examined whether aT_EXO-primed CTL exhibited any other functional traits attributed to typical Tm cells. These traits include 1) secretion of IFN-γ upon Ag stimulation, 2) the enhanced survival and proliferation in response to IL-7 (24), and 3) the capacity to generate Ag-specific CTL. Our data showed that both DC_OVA- and aT_EXO-primed CD8⁺ T cells secrete IFN-γ upon Ag stimulation by EG7 tumor cells expressing OVA (Fig. 4d). However, aT_EXO-primed CTL expanded better in the presence of IL-2 and IL-7 than DC_OVA-primed ones (Fig. 4e). In a chromium release assay, aT_EXO-primed CTL (aT_EXO/OT I_6.1) showed cytotoxicity to EG7 tumor cells, but at a relatively lower level than DC_OVA-primed ones (DC_OVA/OT I_6.1) (Fig. 4f). In addition, the nontreated OT I_6.1 cells used as a control did not show any killing activity to EG7 tumor cells. Taken together, our results indicate that DC_OVA-primed CD44⁺CD62L^lowIL-7R⁻and aT_EXO-primed CD44⁺CD62L^highIL-7R⁺ CTL, which have high and low cytotoxicity to tumor cells, are consistent with typical effector and central memory CTL (emCTL and cmCTL), respectively (12, 25).

CD4⁺ aT_EXO cells stimulate in vivo CD4⁺ T cell-independent CD8⁺ T cell proliferation

We then performed a tetramer staining assay to detect OVA-specific CD8⁺ T cells in wild-type C57BL/6 (B6) or MHC class II (Ia^b) gene KO mice. As shown in Fig. 5a, DC_OVA and aT_EXO cells stimulated proliferation of H-2K^b /OVA_257–264 tetramer-positive CD8⁺ T cells accounting for 1.03 and 2.24% of the total spleen CD8⁺ T cell population in B6 mice, respectively, indicating that aT_EXO is a significantly stronger stimulator than DC_OVA ( p < 0.05). In Ia^b gene KO mice lacking CD4⁺ T cells, however, only aT_EXO cells (2.01%), but not DC_OVA, could stimulate OVA-specific CD8⁺ T cell responses, indicating that the aT_EXO-induced CD8⁺ T cell response is CD4⁺ T cell independent, whereas those of DC_OVA are CD4⁺ T cell dependent.

FIGURE 5.

Stimulation of CD8⁺ T cell proliferation and differentiation into CTL in vivo. Wild-type C57BL/6 or Ia^b−/− gene KO mice were i.v. immunized with irradiated (a) DC_OVA, aT and aT_EXO, and (b) aT_EXO with various gene KO, respectively. Six days after immunization, the tail blood samples of immunized mice were incubated with PE-H-2K^b/OVAI tetramer and FITC-anti-CD8 Ab (Beckman Coulter) according to the company’s protocol, then analyzed by flow cytometry. The value in each panel represents the percentage of tetramer-positive CD8⁺ T cells vs the total CD8⁺ T cell population. The value in parentheses represent the SD. *, p < 0.05 vs cohorts of the positive control (aT/_EXO) (Student’s t test). c, In an in vivo cytotoxicity assay, the above immunized mice were i.v. coinjected at a 1:1 ratio of splenocytes labeled with high (3.0 μM, CFSE^high) and low (0.6 μM, CFSE^low) concentrations of CFSE and pulsed with OVAI and Mut 1 peptide, respectively, 6 days after immunization with aT, aT_EXO, and aT_EXO with various gene KO, respectively. Sixteen hours after target cell delivery, the residual CFSE^high and CFSE^low target cells remaining in the recipients’ spleens were sorted and analyzed by flow cytometry. The value in each panel represents the percentage of CFSE^high cells vs CFSE^low cells remaining in the spleens. *, p < 0.05 vs cohorts of the positive control (aT/EXO_CFSE) (Student’s t test). One representative experiment of three in the above different experiments is shown.

CD4⁺ aT_EXO cell-induced CTL responses and antitumor immunity are mediated by IL-2 and acquired exosomal CD80 specifically delivered via acquired exosomal pMHC I

To elucidate the immune mechanism involved in aT_EXO-induced CTL responses, we performed animal studies by using aT_EXO with different gene KO. The stimulation of OVA-specific CD8⁺ T cell responses by aT_EXO(CD54^−/−) was reduced to more than half (1.08%) of the original one by aT_EXO (2.24%) (Fig. 5b), possibly due to lacking CD54/LFA-1 interaction leading to decreased EXO uptake by aT_EXO(CD54^−/−) cells. Interestingly, the stimulation of OVA-specific CD8⁺ T cell responses by aT_EXO(IL-2^−/−) (0.24%) and aT_EXO(CD80^−/−) (0.42%) cells, but not by aT_EXO(IFN-γ ^−/−) (2.15%), aT_EXO(TNF-α ^−/−) (2.13%) and aT_EXO(CD40^−/−) (2.21%), was greatly lost and significantly less than that by aT_EXO ( p < 0.05), indicating that the aT_EXO stimulatory effect is mediated by its IL-2 and acquired CD80. Furthermore, aT_EXO(K ^b−/−) cells (0.11%) with similar cytokine profiles as aT_EXO, but without acquired pMHC I complexes, also completely lost their stimulatory effect, indicating that the aT_EXO stimulatory effect is specifically delivered to CD8⁺ T cells in vivo via acquired exosomal pMHC I complexes. To assess aT_EXO-induced CD8⁺ T cell differentiation into CTL, we adoptively transferred OVAI peptide-pulsed splenocytes that had been strongly labeled with CFSE (CFSE^high), as well as the control peptide Mut 1-pulsed splenocytes that had been weakly labeled with CFSE (CFSE^low), into the recipient mice that had been vaccinated with DC_OVA and CD4⁺ aT_EXO cells, respectively. As expected, the mice immunized with aT_EXO had larger loss of CFSE ^high cells than DC_OVA (75%) and CD4⁺ aT_EXO cells (88%) (Fig. 5c), indicating that CD4⁺ aT_EXO cells can most efficiently stimulate CD8⁺ T cell differentiation into CTL effectors. Interestingly, the OVA-specific cytotoxicity was substantially lost in CD4⁺ aT_EXO(IL-2^−/−)-(2%) and aT_EXO (CD80^−/−)-immunized (15%) mice and significantly less than the CD4⁺ aT_EXO-induced one (p < 0.05), but not in CD4⁺ aT_EXO (IFN-γ ^−/−)-(89%), aT_EXO(TNF-α ^−/−)-(90%), and aT_EXO (CD40^−/−)-immunized (88%) mice, thus further confirming that CD4⁺ the aT_EXO stimulatory effect is mediated by its IL-2 secretion and acquired CD80 costimulation. In addition, the CD4⁺ aT_EXO(K ^b−/−)-vaccinated mice did not display any killing activity (3%), again confirming that the acquired pMHC I complexes play a critical role in targeting the CD4⁺ aT_EXO stimulatory effect to OVA-specific CD8⁺ T cells in vivo. Furthermore, the results derived from our in vivo antitumor immunity studies are also consistent with the above aT_EXO-induced in vivo CD8⁺ CTL responses (Fig. 5a). Our data showed that 1) CD4⁺ aT_EXO cells induced stronger antitumor immunity than DC_OVA (Expt. I of Table I), and 2) IL-2 and acquired exosomal CD80 of CD4⁺ aT_EXO are specifically delivered to CD8⁺ T cells in vivo via acquired exosomal pMHC I (Expt. II of Table I).

Table I.

EXO-targeted CD4⁺ T cell vaccine protects against lung tumor metastases

Vaccinesa	Tumor Growth Incidence (%)	Median Number of Lung Tumor Colonies
Expt. 1
DCOVA	2/8 (25)	17 ± 6
aTEXO	0/8 (0)	0
aT	8/8 (100)	>100
PBS	8/8 (100)	>100
Expt. 2
aTEXO(IL-2−/−)	7/8 (88)	78 ± 20
aTEXO(IFN-γ −/−)	0/8 (0)	0
aTEXO (TNF-α −/−)	0/8 (0)	0
aTEXO (CD40−/−)	0/8 (0)	0
aTEXO (CD54−/−)	3/8 (38)	26 ± 8
aTEXO (CD80−/−)	5/8 (63)	47 ± 23
aTEXO (K b−/−)	7/8 (88)	84 ± 17
PBS	8/8 (100)	>100
Expt. 3
DCOVA	4/8 (50)	14 ± 9
aTEXO	0/8 (0)	0
PBS	8/8 (100)	>100
Expt. 4
DCOVA	2/8 (25)	17 ± 6
DCOVA plus Tr cells	8/8 (100)	>100
aTEXO	0/8 (0)	0
aTEXO plus Tr cells	0/8 (0)	0
PBS	8/8 (100)	>100

^aIn experiment 1, the wild-type C57BL/6 (B6) mice (n = 8) were i.v. immunized with DC_OVA, aT and aT_EXO cells, or PBS. In experiment 2, the wild-type C57BL/6 mice (n = 8) were i.v. immunized with either aT_EXO cells or aT_EXO cells with various gene KO. Six days after the immunization, each mouse was challenged i.v. with OVA-expressing (BL6–10_OVA) tumor cells (0.5 × 10⁶ cells/mouse). In experiment 3, the wild-type C57BL/6 mice (n = 8) were i.v. immunized with DC_OVA, aT_EXO, or PBS. Three months after the immunization, each mouse was challenged i.v. with BL6–10_OVA tumor cells (2 × 10⁶ cells/mouse). In experiment 4, the wild-type C57BL/6 mice (n = 8) were i.v. immunized with DC_OVA, aT_EXO cells alone or together with CD4⁺25⁺ Tr cells. Six days after the immunization, each mouse was challenged i.v. with OVA-expressing (BL6–10_OVA) tumor cells (0.5 × 10⁶ cells/ mouse). All the mice were sacrificed 4 wk after tumor cell challenge and the numbers of lung metastatic tumor colonies were counted. One representative experiment of three is shown.

CD4⁺ aT_EXO cells induce efficient long-term OVA-specific CD8⁺ T cell memory

Active CD8⁺ T cells can become long-lived Tm cells after adoptive transfer in vivo (26). We then assessed whether these CD4⁺ aT_EXO-primed CD8⁺T cells can also become long-lived Tm cells. As shown in Fig. 6a, we still detected 0.46% OVA-specific CD8⁺ T cells in peripheral blood of mice immunized with aT_EXO, which is significantly stronger than that (0.12%) in mice immunized with DC_OVA 3 mo after the immunization ( p < 0.05). These OVA-specific CD8⁺ T cells were also CD44⁺, indicating that they are OVA-specific CD8⁺ Tm cells. In addition, the survived CD4⁺ aT_EXO-primed CD8⁺ Tm cells are nearly 4-fold compared with the survived DC_OVA-stimulated ones, indicating that CD4⁺ aT_EXO-primed CD44⁺CD62L^highIL-7R⁺ CTL with low cytotoxicity to tumor cells are long survival cmCTL. The recall responses were then assessed on day 4 after the boost of immunized mice with DC_OVA. As expected, CD8⁺ Tm cells were expanded by 10-fold in these immunized mice after the boost (Fig. 6b), indicating that these CD8⁺ Tm cells are functional. In another set of experiments, the above immunized mice were challenged with a high dose (2 × 10⁶ cells/mouse) of BL6–10_OVA tumor cells. Only four of eight (50%) mice immunized with DC_OVA were tumor free, whereas all eight of eight (100%) mice immunized with CD4⁺ aT_EXO cells did not have any lung metastasis (Expt. III of Table I), indicating that CD4⁺ aT_EXO cells can induce more efficient long-term CD8⁺ T cell memory than DC_OVA.

FIGURE 6.

Development of Ag-specific CD8⁺ memory T cells. a, C57BL/6 mice were immunized with irradiated DC_OVA and aT_EXO, respectively. Three months later, the tail blood was taken from these immunized mice and stained with PE-H-2K^b/OVA tetramer, FITC-anti-CD8 and ECD-anti-CD44 Abs, and analyzed by flow cytometry. The value in each panel represents the percentage of tetramer-positive CD8⁺ T cells vs the total CD8⁺ T cell population. The value in parentheses represents the SD. The PE-tetramer- and FITC-CD8-positive cells in the squares were sorted and analyzed, showing they were also PE-tetramer- and ECD-CD44-positive cells in the circles. *, p < 0.05 vs cohorts of the positive control (DC_OVA) (Student’s t test). b, The above immunized mice were boosted with DC_OVA. Four days after the boost, the recall responses were examined using staining with PE-H-2K^b/OVA tetramer and FITC-anti-CD8 Ab and analyzed by flow cytometry. The value in each panel represents the percentage of tetramer-positive CD8⁺ T cells vs the total CD8⁺ T cell population. The value in parentheses represents the SD. The results presented are representative of five separate mice per group. One representative experiment of three is shown.

CD4⁺ aT_EXO cells counteract in vivo CD4⁺25⁺ Tr cell-mediated suppression of OVA-specific CD8⁺ T cell proliferation and antitumor immunity

Because we demonstrated above the counteraction of in vitro CD4⁺25⁺ Tr cell-mediated suppression, we wanted to further assess the potential effect of CD4⁺ aT_EXO cells in counteraction of in vivo CD4⁺25⁺ Tr cell-mediated suppression. We performed in vivo proliferation and cytotoxicity assays as described above. As shown in Fig. 7a, the stimulation of OVA-specific CD8⁺ T cell responses in mice by DC_OVA (1.03%) was almost completely lost (0.10%) when mice were coinjected with CD4⁺25⁺ Tr cells ( p < 0.05), indicating that CD4⁺25⁺ Tr cells can significantly inhibit DC_OVA-stimulated CD8⁺T cell responses. However, the OVA-specific CD8⁺ T cell responses in mice with CD4⁺ aT_EXO stimulation in presence or absence of CD4⁺25⁺ Tr cells maintained at a similar level (2.24%). In the in vivo cytotoxicity assay, the effector CTL responses derived from DC_OVA, but not CD4⁺ aT_EXO cells, in presence of CD4⁺25⁺ Tr cells were lost (Fig. 7b). Furthermore, the results derived from our in vivo antitumor immunity studies (Expt. IV of Table I) are also consistent with the above aT_EXO-induced in vivo CD8⁺ CTL responses (Fig. 5). Our data demonstrated that CD4⁺ aT_EXO cells, but not DC_OVA, induced efficient antitumor immunity in mice even with coinjection of CD4⁺25⁺ Tr cells, which was against the challenge of BL6–10_OVA tumor cells from developing lung tumor metastasis (Expt. IV of Table I), indicating that CD4⁺ aT_EXO cells can counteract in vivo CD4⁺25⁺ Tr cell-mediated immune suppression, possibly through bypassing the CD4⁺25⁺ Tr cell-mediated suppressive pathway.

FIGURE 7.

CD4⁺25⁺ Tr-mediated suppression of CD8⁺ T cell proliferation and differentiation in vivo. Wild-type C57BL/6 mice were i.v. immunized with irradiated (a) DC_OVA, aT and aT_EXO, alone or with CD4⁺25⁺ Tr cells, respectively. Six days after immunization, the tail blood samples of immunized mice were incubated with PE-H-2K^b/OVAI tetramer and FITC-anti-CD8 Ab according to the company’s protocol, then analyzed by flow cytometry. The value in each panel represents the percentage of tetramer-positive CD8⁺ T cells vs the total CD8⁺ T cell population. The value in parentheses represents the SD. *, p < 0.05 vs cohorts of the positive control (aT_EXO in absence of Tr cells) (Student’s t test). b, In an in vivo cytotoxicity assay, the above immunized mice were i.v. coinjected at a 1:1 ratio of splenocytes labeled with high (3.0 μM, CFSE^high) and low (0.6 μM, CFSE^low) concentrations of CFSE and pulsed with OVAI and Mut 1 peptide, respectively, 6 days after immunization. Sixteen hours after target cell delivery, the residual CFSE^high and CFSE^low target cells remaining in the recipients’ spleens were sorted and analyzed by flow cytometry. The value in each panel represents the percentage of CFSE^high cells vs CFSE^low cells remaining in the spleens. *, p < 0.05 vs cohorts of the positive control (aT_EXO in absence of Tr cells) (Student’s t test). One representative experiment of three in the above different experiments is shown.

Discussion

According to the progressive linear differentiation hypothesis (27), T cell differentiation involves a phase of proliferation preceding the acquisition of fitness and effector function. Primed CD8⁺ T cells reach a variety of differentiation stages that contain effector cells as well as cells that have been arrested at intermediate levels of differentiation. Thus, they retain a flexible gene imprinting. T cells that may survive after the retraction phase of an immune response can be resolved into distinct subsets of either cmCTLs representing cells at intermediate levels of differentiation or fully differentiated emCTLs with effector capacity (28, 29). It has been shown that a strong Ag presentation stimulates development of effector CTL, whereas a less efficient Ag presentation can lead to generation of central memory CTL (30). In this study, we demonstrated that CD4⁺ aT_EXO cells were able to stimulate naive CD8⁺ T cell differentiation into CD8⁺44⁺62^highIL-7R⁺ cmCTLs with less cytotoxicity and longer survival capacity leading to strong memory T cell responses, compared with DC_OVA-primed CD8⁺44⁺62^lowIL-7R⁺ emCTLs with high cytotoxicity and shorter survival capacity in vivo. Therefore, an EXO-targeted CD4⁺ T cell vaccine using peripheral blood CD4⁺ T cells of a cancer patient incubated with tumor cell-derived EXO from ascites of a cancer patient may be a useful strategy for EXO-based treatment of cancer.

Administration of attenuated T lymphocytes to animals has been shown to stimulate immune suppression and to prevent the development of experimental autoimmune diseases (31, 32). Vaccination using myelin basic protein autoreactive T cells has also been applied to clinical trial in multiple sclerosis (33). In this study, we clearly showed that CD4⁺ aT_EXO cells can more strongly stimulate OVA-specific immunogenic CD8⁺ CTL responses, antitumor immunity, and CD8⁺ T cell memory in wild-type mice than EXO and DC_OVA. Interestingly, the CD8⁺CTL responses stimulated by are found to be CD4⁺T cell independent. Thus, EXO-aT_EXO targeted T cell vaccines may be very useful in induction of anti-HIV immunity in HIV patients with CD4⁺ T cell deficiency because HIV-1 infection is characterized by a gradual loss of CD4⁺ T cells and progressive immune deficiency (34). Furthermore, we also elucidated the molecular mechanism involved in aT_EXO vaccines, which includes 1) the IL-2 secretion and the acquired exosomal CD80 costimulation that mediate the aT_EXO stimulatory effect and 2) the acquired exosomal pMHC I complexes that play a critical role in targeting the aT_EXO stimulatory effect to CD8⁺ T cells in vivo.

CD4⁺25⁺ Tr cells develop in the thymus and then enter peripheral tissues where they suppress activation of other self-reactive T cells (35, 36). It has been reported that an elevated number of Tr cells was detected in tumors (37, 38) which suppressed the anti-tumor immune responses by inhibition of CD4⁺T cell proliferation and a helper effect (39–41) as well as DC maturation (42). Therefore, how to combat immune tolerance becomes a critical challenge in cancer immunotherapy (43). A variety of stimuli that increase the potency of T cell stimulation have been shown to abrogate CD4⁺ Tr cell function including high Ag dose, TLR signals, CD28 engagement, and provision of IL-2 (44–48). In this study, we demonstrated that CD4⁺ aT_EXO cells, but not DC_OVA, can stimulate CD8⁺ T cell proliferation in the presence of CD4⁺25⁺ Tr cells in vitro and in vivo. These CD4⁺ aT_EXO cells expressing exosomal pMHC I and CD80 and secreting IL-2 can break Tr cell-mediated immune tolerance, possibly due to its capacity to 1) directly stimulate CD8⁺ T cell responses in absence of CD4⁺ Th cells and DC, thus bypassing the above Tr cell-mediated suppressive pathways or 2) directly counteract CD4⁺ Tr suppression via CD28 engagement and IL-2 stimulation (44–48).

Taken together, our data show that DC_OVA-derived EXO_OVA can be uptaken by CD4⁺ T cells. EXO_OVA-targeted CD4⁺ aT_EXO cells expressing acquired exosomal pMHC I and CD80 can stimulate CD8⁺ cmCTL responses, more efficient in vivo antitumor immunity and long-term CD8⁺ T cell memory than DC_OVA, and counteract CD4⁺25⁺ Tr cell-mediated suppression of OVA-specific CTL responses and antitumor immunity. The aT_EXO stimulatory effect is mediated via its IL-2 secretion and acquired exosomal CD80 costimulation and is specifically targeted to CD8⁺T cells in vivo via acquired exosomal pMHC I complexes. Therefore, an EXO-targeted active CD4⁺ T cell vaccine may represent a novel and highly effective vaccine strategy for inducing immune responses not only against tumors, but also other infectious diseases.