Normothermic liver perfusion derived extracellular vesicles have concentration-dependent immunoregulatory properties

Abstract

Extracellular vesicles (EVs) are major contributors to immunological responses following solid organ transplantation. Donor derived EVs are best known for their role in transplant rejection through transferring donor major histocompatibility complex proteins to recipient antigen presenting cells, a phenomenon known as ‛cross-decoration'. In contrast, donor liver-derived EVs are associated with organ tolerance in small animal models. Therefore, the cellular source of EVs and their cargo could influence their downstream immunological effects. To investigate the immunological effects of EVs released by the liver in a physiological and transplant-relevant model, we isolated EVs being produced during normothermic ex vivo liver perfusion (NEVLP), a novel method of liver storage prior to transplantation. We found EVs were produced by the liver during NEVLP, and these EVs contained multiple anti-inflammatory miRNA species. In terms of function, liver-derived EVs were able to cross-decorate allogeneic cells and suppress the immune response in allogeneic mixed lymphocyte reactions in a concentration-dependent fashion. In terms of cytokine response, the addition of 1 × 10⁹ EVs to the mixed lymphocyte reactions significantly decreased the production of the inflammatory cytokines TNF-α, IL-10 and IFN-γ. In conclusion, we determined physiologically produced liver-derived EVs are immunologically regulatory, which has implications for their role and potential modification in solid organ transplantation.

Keywords: anti‐inflammatory; cross‐decoration; extracellular vesicles; liver‐derived; normothermic liver perfusion; organ tolerance; organ transplantation.

FIGURE 1

Experimental schematic. The schematic of the experimental procedures in this study. A rat liver produced EVs during NEVLP. EVs were isolated from samples taken every hour during the perfusion and from the remaining perfusate at the end of the perfusion. The isolated EVs were characterised and tested for functionality.

FIGURE 2

Characterisation of EVs produced by the liver during NEVLP. (a) Representative cryogenic electron microscopy images of the perfusate samples indicate the presence and the expected morphology of the EVs at T4 as well as the lack of EVs at T0. Scalebar: 100 nm; inset 50 nm. (b) The purified liver EVs and liver cell lysates (5–20 µg) were loaded to detect EV markers, TSG101 and CD63 via western blot analysis. (c) Representative nanosight tracking analysis (NTA) data showing the size and concentration of EV isolates from their respective groups, D) the mode size distribution of EVs isolated from each group and (e) the particle concentrations at T0, T1 and T4, indicating that the concentration of EVs throughout the perfusion.

FIGURE 3

Liver‐derived EVs contain multiple anti‐inflammatory miRNAs. (a) The MDS plot indicates that the levels of miRNAs isolated from the T0 samples differ from those of the T4 samples. (b) The heat map indicates the changes in miRNA expression between the two groups and (c) the volcano plot indicates the miRNAs most upregulated over the course of the perfusion.

FIGURE 4

miRNA pathway enrichment. (a) KEGG pathway enrichment and (b) GO‐BP pathway enrichment for differentially expressed miRNAs between the start and the end of perfusion.

FIGURE 5

Cross‐decorating efficiency of liver‐derived EVs is concentration dependent. (a) Representative flow cytometry plots of BN splenocytes with the addition of either BN or various concentrations of LEW liver‐derived EVs. The antibody clones B5 and OX3 are specific for LEW class I and II major histocompatibility complexes, respectively. The BN APC gating strategy is shown in Figure S2. (b) Bar graph summary of each quadrant of the flow plots in A, with data being presented as percent change relative to 0 EV control following background subtraction of negative (BN EV) control.

FIGURE 6

Liver‐derived EVs can decrease an allogenic proliferation response. (a) Relative proliferation of BN lymph node cells stimulated with irradiated LEW splenocytes 4:1 and treated with LEW EVs at various concentrations. The non‐specific proliferation response (positive control) showed that all cells are capable of proliferation (data not shown). (b) MLR supernatant was collected and cytokines were measured. Data are presented as percent change relative to the 0 EV control following background subtraction of negative control. Asterisks indicate p values as follows: *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.