Immune engineered extracellular vesicles to modulate T cell activation in the context of type 1 diabetes

Extracellular vesicles (EVs) can affect immune responses through antigen presentation and costimulation or coinhibition. We generated designer EVs to modulate T cells in the context of type 1 diabetes, a T cell–mediated autoimmune disease, by engineering a lymphoblast cell line, K562, to express HLA-A*02 (HLA-A2) alongside costimulatory CD80 and/or coinhibitory programmed death ligand 1 (PD-L1). EVs presenting HLA-A2 and CD80 activated CD8+ T cells in a dose, antigen, and HLA-specific manner. Adding PD-L1 to these EVs produced an immunoregulatory response, reducing CD8+ T cell activation and cytotoxicity in vitro. EVs alone could not stimulate T cells without antigen-presenting cells. EVs lacking CD80 were ineffective at modulating CD8+ T cell activation, suggesting that both peptide-HLA complex and costimulation are required for EV-mediated immune modulation. These results provide mechanistic insight into the rational design of EVs as a cell-free approach to immunotherapy that can be tailored to promote inflammatory or tolerogenic immune responses.

This PDF file includes:

Fig. S4. Reproducibility of EV isolations using chromatography combined with ultrafiltration. (A)
NTA histograms from eight independent EV isolations using BE-SEC combined with ultrafiltration, showing consistent batch-to-batch particle sizes ranging from 42-75 nm.(B) UV chromatograms from the same eight EV isolations, showing good peak separation of EVs from soluble proteins that is consistent between batches of different starting volumes.(C) Western blot analysis of EV isolates from ultracentrifugation (UC) or BE-SEC combined with ultrafiltration, showing the presence of the canonical small EV marker ALIX in both samples.(D) NTA histograms from six independent EV isolations using ultracentrifugation, showing wider batchto-batch variability and less consistent particle size distributions compared to BE-SEC isolations.

Fig
Fig. S3.Characterization of THP-1 EVs.(A) NTA showing representative particle concentration and size distribution of EVs derived from THP-1 cells.(B) Western blot analysis of EV isolates demonstrating the presence of canonical EV markers.Representative of 4 independent experiments.

Fig
Fig. S5.Confirming antigen-driven activation of J-1E6 T cells in co-cultures.J-1E6 cells cultured with K562 A2/PPI cells and exogenous PPI15-24 have increased IL-2 secretion compared to cultures without exogenous peptide, whereas culture with IGRP265-273 peptide does not increase IL-2 secretion.Statistical differences for were determined by one-way ANOVA followed by Tukey's multiple comparisons test.* p < 0.05, **** p < 0.0001.
Fig S6.Effects of THP-1 EVs on J-1E6 T cell activation.IL-2 secretion from J-1E6 T cells after coculture with K562 A2/PPI cells and increasing amounts of T-EVs, showing no significant change in T cell activation with increasing amounts of EVs.n = 3. Statistical differences were determined by one-way ANOVA followed by Tukey's multiple comparison test.ns = nonsignificant.
Fig S8.Driving PD-L1 expression in THP-1 cells and EVs, and effects on J-1E6 T cell activation.(A) Surface expression of PD-L1 in THP-1 cells with or without IFN-γ treatment.n = 3. (B) ELISA of PD-L1 on EVs from THP-1 cells with or without IFN-γ treatment compared to total EV protein content.n = 3. (C) IL-2 secretion from J-1E6 T cells after co-culture with K562 cells and THP-1 EVs with or without PD-L1, showing slightly decreased T cell activation with excessive amounts of T-EVs or moderate amounts of T-PD-L1-EVs.Statistical differences for (B) were determined by an unpaired, two-tailed t-test.Statistical differences for (C) were determined by one-way ANOVA followed by Tukey's multiple comparison test.* p < 0.05, *** p < 0.001, **** p < 0.0001.

Table S1 .
Antibody information.Antibodies used for flow cytometry and western blot analysis.