Fig 2: The proportion of CD3+ T cells, CD3+CD4-CD8- T cells, CD19+ B cells, CD19-CD138+ cells (Plasma cells), CD3+S1PR1+ T cells and CD3+CD4+S1PR1+ T cells in spleen and early apoptotic cells in CD4+ and CD8+ T cells in lymph nodes were significantly increased in B6-Mir223−/−Faslpr/lpr mice. (A) Various cellular subsets in lymph nodes and spleen. CD3+CD4+ (CD4 T cells), CD3+CD8+ (CD8 T cells), CD3+ CD4−CD8− (Double negative (DN) T cells), CD19+ (B cells), CD19−CD138+ (Plasma cells), CD19+CD138+ (Plasmablasts), CD4+CD44+CD62L+ (Memory T cells), CD4+CD44+CD62L− (Effector T cells), CD69+CD4+ cells, S1PR1+CD3+ cells, S1PR1+CD4+ cells. Distribution of subsets in total cells isolated from whole cervical lymph nodes and spleen were indicated in B6-Mir223−/−Faslpr/lpr (n=7) and B6-Mir223+/+Faslpr/lpr (n = 8) mice. Absolute number of total splenocyte and lymph nodes cell had no difference between two genotypes (Supplementary Figure 2E). (B) Apoptotic cells in lymph nodes and spleen. Annexin V+7-AAD− (Early apoptosis cells), Annexin V+7-AAD+ (Late apoptotic cells). Distribution of subsets in CD3+CD4+ T cells or CD3+CD8+ T cells isolated from whole cervical lymph nodes and spleen were indicated in B6-Mir223−/−Faslpr/lpr (n = 3) and B6-Mir223+/+Faslpr/lpr (n = 3) mice. *p < 0.05, **p < 0.01, ns, not significant, by Student’s t-test in (A), by Wilcoxon signed-rank test in (B).

Fig 3: Enhanced ex-vivo cytotoxicity of cells isolated from TDLNs of YUMM1.7-CM compared to YUMM1.7 tumor-bearing mice. (A) Representative bright field (BF)/fluorescence micrographs of EdU-prelabeled YUMM1.7 and YUMM1.7-CM cells cultured alone (top) or co-cultured with immune cells isolated from respective TDLNs (bottom). EdU positive nuclei are observed in green. (B) Quantitation of YUMM1.7 and YUMM1.7-CM cells elimination following co-incubation with TDLNs cells measured by in-well EdU fluorescence readings or by EdU positive cell counts, as indicated. Data are presented as mean ± SEM obtained from four independent YUMM1.7-CM or YUMM1.7 cultures. Unpaired Student’s two-tailed t-test; *indicates different from YUMM1.7 co-cultures p < 0.05. (C) Representative micrographs of immune cells surrounding a tumor cell visualized by merged CD8 immunofluorescence (red)/BF images. EdU positive nuclei of YUMM1.7 or YUMM1.7-CM cells are observed in green; nuclei of TDLNs derived effector cells stain blue with DAPI. Scale bars, 10 µm.

Fig 4: AAV2/8-insulin-based therapy elicits a cellular immune response against the vector and the transgene in NOD diabetic mice. Livers of NOD mice treated with AAV2/8-HLP-hINSco (indicated with AAV2/8) show signs of T cell infiltration observed in pre-diabetic NOD mouse pancreata. a Pancreas sections from a 8-week-old control NOD female pancreas (a, top panel), a 30-day AAV2/8-insulin-treated C57BL/6 (a, middle panel) and a 30-day AAV2/8-insulin-treated NOD (a, bottom panel) were stained for insulin (red) and glucagon (green), and CD3 (yellow) to show β-cell destruction in treated mice. b Livers from AAV2/8-HLP-hINSco -injected C57BL/6 (b, left) and NOD (b, right) mice were stained for insulin and CD3 to test for T cell infiltration. c Livers from AAV2/8-HLP-hINSco-injected NOD mice were stained for insulin and CD8. Blue shows nuclear DAPI staining. Scale bar 50 μm. d, e Ex vivo stimulated splenocytes from AAV2/8-HLP-hINSco -treated NOD mice and C57BL/6 controls for IFN-γ ELISPOT assay. Cells from NOD and C57BL/6 mice were harvested after 30 days (d) or more than 200 days (e) from the day of the injection of the AAV2/8-insulin vector. Cells were stimulated in vitro for 40 h with CD8+ immunodominant peptides (either InsB15–23 peptide for NOD or insulin Kb-restricted epitope A12–21 containing peptide for C57BL/6) or the AAV8 capsid-specific CD8+ T cell peptide NSLANPGIA and the number of resulting IFN-γ spots was counted with an ELISPOT reader. f Levels of anti-inflammatory cytokines in NOD mice compared to C57BL/6. Splenocytes from C57BL/6 and NOD mice were harvested at the end of a 30-day treatment with AAV2/8-HLP-hINSco 5 × 109 vg and stimulated with PMA/Iono for 48 h. Supernatants were then collected and tested for IL-10 production. g Mouse anti-AAV8 ELISA performed on blood plasma of diabetic, healthy and 5 × 109 vg AAV2/8-HLP-hINSco-treated NOD and C57BL/6 mice. Treated mice were tested for anti-AAV8 antibodies 30 days after the beginning of the therapy. The absorbances at 405 nm correlate with the concentration of the antibody. *(p ≤ 0.05), **(p ≤ 0.01), ***(p ≤ 0.001), unpaired Student’s t-test. Data shown are expressed as mean ± SE and are representative of two independent experiments

Fig 5: CD4+/CD8a+ cell numbers increase in tumors by SART3/CD40L+GM-CSF vaccination and depletion deteriorates the anti-tumor efficacy.(A) Tissue sections of subcutaneous CT26 tumors after administration of polyplex micelles were immunostained with anti-CD4 or CD8a antibodies. CD4+ and CD8a+ T cells were highly infiltrated into tumors of the SART3/CD40L+GM-CSF vaccine group compared with that in the mock control. *P<0.05 (n = 4). (B) Depletion of CD4+ and/or CD8a+ T cells by i.p. administration of their neutralizing antibodies was confirmed by flow cytometric analysis of blood samples. (C) Kaplan-Meier analysis showed that depletion of CD4+ or CD8a+ T cells decreased the survival of the SART3/CD40L+GM-CSF vaccine group compared with that that of the isotype IgG2 control (P = 0.084 and 0.003, respectively, n = 7−8) in mice with peritoneal dissemination of CT26 cells.