Extracorporeal photopheresis reduces the T cell stimulatory capacity of human primary blood conventional dendritic cells type 1

Abstract

Introduction: Extracorporeal photopheresis (ECP) is an immunomodulatory treatment option for different T cell-mediated diseases such as cutaneous T cell lymphoma (CTCL) and chronic graft-versus-host disease (GvHD). While in CTCL the polarization of T cells is shifted towards T helper cells type 1 (TH1) and an immune response against the lymphoma is induced, ECP in GvHD rather leads to the expansion of regulatory T cells (Treg). How ECP regulates the immune response dependent on the underlying disease is still not exactly known. As dendritic cells (DCs) are crucial regulators of the immune response, it is supposed that they are key players in the immunomodulatory effects of ECP. However, due to the scarcity of primary DCs in blood, research has focused on in vitro-generated monocyte-derived DCs so far.

Methods: Here, we present for the first time how the primary human blood DC subpopulations, i.e., conventional DCs type 1 (cDC1), cDC2, DC3, and plasmacytoid DCs (pDC), directly isolated from blood of healthy donors, respond to in vitro ECP treatment.

Results: We demonstrate that the exposure to 8-methoxypsoralen and UV-A light irradiation induces apoptosis in Toll-like receptor ligand-activated cDC1 and pDC as well as - to a minor extent - in steady state cDC1, cDC2, and DC3. However, the selective effect of ECP on viability of DC subpopulations was dependent on culture duration (18h vs. 42h) as well as condition (steady state vs. TLR ligand activated). Further, ECP modulates the expression of the co-stimulatory and co-regulatory molecules CD40, CD86, and PD-L1 on DC subpopulations. While ECP did not affect the T cell stimulatory capacity of cDC2 and DC3, ECP-treated cDC1 and - to a minor extent - pDC showed reduced activation of memory T cells and diminished secretion of TH1- and TH17-associated cytokines.

Conclusion: Thus, especially blood cDC1 are direct targets of ECP and the reduction of their T cell stimulatory capacity might contribute to the clinical efficacy observed in chronic GvHD patients.

Keywords: ECP; GvHD; T cells; apoptosis; cDC1; dendritic cells; extracorporeal photopheresis; graft-versus-host disease.

Figure 1

Applied gating strategy for the isolation of human DC subpopulations. PBMCs of healthy human adults undergoing thrombocytapheresis were enriched using the EasySep Pan-DC Pre-Enrichment kit and stained with a panel of fluorochrome-conjugated antibodies ( Table 1 ). Then, DC subpopulations were sorted using a BD Aria II by gating for the morphology of leukocytes (FSC-A/SSC-A), singlets (FSC-A/FSH-H), and living cells (DAPI^-). After exclusion of T cells (CD3⁺), B cells (CD19/CD20⁺), NK cells (CD56/CD335⁺), and monocytes (CD14/CD16⁺), DCs were selected by gating for HLA-DR⁺ cells. DCs were divided into CD1c⁺CD123^-, CD1c^-CD123^-, and CD1c^-CD123⁺ cells using a quadrant gate. CD1c⁺ DCs were identified in CD1c⁺CD123^- cells by co-expression of CD1c and CD11c and sorted into CD64^-CD163^- cDC2 (red) and CD64⁺CD163⁺ DC3 (purple). cDC1 were gated in the CD1c^-CD123^- fraction as CD141⁺CD11c^int cells (yellow-orange). In the CD1c^-CD123⁺ quadrant, pDC were sorted as CD123⁺CD141^int cells. One representative donor is shown.

Figure 2

Experimental ECP induces apoptosis primarily in human blood cDC1 after 18 h of culture. Cell sorter-purified cDC1, cDC2, CD14^- DC3, and pDC were incubated either with 400 ng/ml 8-MOP or equal amount of solvent control (ethanol) for 30 min at 37°C as indicated below the figure. Then, cells were either irradiated with 2 J/cm² UV-A light or mock-treated. After centrifugation to remove the solvent, cells were resuspended in (A) medium or (B) medium containing 5 µg/ml R848. After 18 h of culture, DCs were stained with the antibodies used for cell sorting and 7-AAD and Annexin V-PE to determine viability. Truncated violin plots depict percentages of alive (Annexin V^-/7-AAD^-), early apoptotic (Annexin V⁺/7-AAD^-), late apoptotic (Annexin V⁺/7-AAD⁺), and necrotic (Annexin V^-/7-AAD⁺) cDC1 (yellow-orange symbols), cDC2 (red symbols), DC3 (purple symbols) and pDC (blue symbols) of six donors (each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 3

Experimental ECP modulates the expression of co-stimulatory and -regulatory molecules on human primary DCs after 18 h of culture. Cell sorter-purified cDC1, cDC2, CD14^- DC3, and pDC were incubated either with 400 ng/ml 8-MOP or equal amount of solvent control (ethanol) for 30 min at 37°C as indicated below the figure. Then, cells were either irradiated with 2 J/cm² UV-A light or mock-treated. After centrifugation to remove the solvent, cells were resuspended in (A) medium or (B) medium containing 5 µg/ml R848. After 18 h of culture, DCs were stained with the antibodies used for cell sorting and A700-coupled anti-CD40, FITC-coupled anti-CD86, and BV650-coupled anti-PD-L1 or respective isotype controls. Truncated violin plots show ΔMFI on alive (Annexin V^-/7-AAD^-) cDC1 (yellow-orange symbols), cDC2 (red symbols), DC3 (purple symbols) and pDC (blue symbols) of six donors (each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 4

R848-induced cyto- and chemokine secretion by DCs is minorily influenced by in vitro treatment with ECP. Supernatants of DC subpopulations treated either with ECP (8-MOP/UV-A) or control conditions for (A) 18 h or (B) 42 h were analyzed by LEGENDplex Human Macrophage/Microglia Panel (BioLegend) for the secretion of cyto- and chemokines. Measured concentrations (pg/ml) were normalized based on the highest measured median value for each analyte in the whole data set. Percentage of maximum for cDC1, cDC2, CD14^- DC3, pDC is shown (each square shows the mean of (A) six or (B) five donors).

Figure 5

Experimental ECP induces apoptosis primarily in human blood cDC1 and pDC after 42 h of culture. Cell sorter-purified cDC1, cDC2, CD14^- DC3, and pDC were incubated either with 400 ng/ml 8-MOP or equal amount of solvent control (ethanol) for 30 min at 37°C as indicated below the figure. Then, cells were either irradiated with 2 J/cm² UV-A light or mock-treated. After centrifugation to remove the solvent, cells were resuspended in (A) medium or (B) medium containing 5 µg/ml R848. After 42 h of culture, DCs were stained with the antibodies used for cell sorting and 7-AAD and Annexin V-PE to determine viability. Truncated violin plots depict percentages of alive (Annexin V^-/7-AAD^-), early apoptotic (Annexin V⁺/7-AAD^-), late apoptotic (Annexin V⁺/7-AAD⁺), and necrotic (Annexin V^-/7-AAD⁺) cDC1 (yellow-orange symbols), cDC2 (red symbols), DC3 (purple symbols) and pDC (blue symbols) of five donors (each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 6

Experimental ECP induces minor changes in the expression of co-stimulatory and -regulatory molecules on human primary DCs after 42 h of culture. Cell sorter-purified cDC1, cDC2, CD14^- DC3, and pDC were incubated either with 400 ng/ml 8-MOP or equal amount of solvent control (ethanol) for 30 min at 37°C as indicated below the figure. Then, cells were either irradiated with 2 J/cm² UV-A light or mock-treated. After centrifugation to remove the solvent, cells were resuspended in (A) medium or (B) medium containing 5 µg/ml R848. After 42 h of culture, DCs were stained with the antibodies used for cell sorting and A700-coupled anti-CD40, FITC-coupled anti-CD86, and BV650-coupled anti-PD-L1 or respective isotype controls. Truncated violin plots show ΔMFI on alive (Annexin V^-/7-AAD^-) cDC1 (yellow-orange symbols), cDC2 (red symbols), DC3 (purple symbols) and pDC (blue symbols) of five donors (each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 7

In vitro ECP of human primary blood cDC1 strongly reduces their capacity to activate memory T cells. Cell-sorted cDC1 (yellow-orange), cDC2 (red), CD14^- DC3 (purple), and pDC (blue) were treated with 8-MOP and UV-A light. After washing, DCs were (A) pulsed with CEFT peptides, (B) pulsed with CEFT peptides in presence of 1 µg/ml R848, or (C) incubated with 10 CFU/DC heat-killed E. coli. After 18 h of culture, autologous CFSE-labelled memory T cells were added (1:10 DC:T cell ratio) and co-cultured for five days. T cells were stained with a panel of fluorochrome-coupled antibodies and acquired using a Cyoflex S (Beckman Coulter). T cells were gated as shown in Supplementary Figure 6 . Truncated violin plots depict percentages of proliferated and activated (CFSE^-CD25⁺) CD4⁺ (left panel) and CD8⁺ (right panel) memory T cells of six donors (cDC1 in (A) five donors; each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).

Figure 8

Experimental ECP of primary blood DCs does not enhance the expression of exhaustion markers or death receptors on activated memory T cells. Proliferated and activated memory T cells from Figure 7 were analyzed for the expression of CD25, CD71, CD178, and CD223 by flow cytometry. Data were normalized to the highest measured value in the data set and the relative values (percentage of maximum) plotted as heatmap. Each square shows the mean of six donors (cDC1 in (A) five donors).

Figure 9

In vitro ECP-treatment of human cDC1 reduces the secretion of TH1- and TH17-associated cytokines by memory T cells. Supernatants from DC:T cell co-cultures shown in Figure 9C were analyzed for the concentration of T cell-associated cytokines using the LEGENDplex Hu Th Cytokine Panel (BioLegend). Truncated violin plots show the concentrations of (A) IL-2, (B) TNFα, (C) IFNγ, (D) IL-22, (E) IL-17A, (F) IL-17F, (G) IL-6, (H) IL-9, (I) IL-10, (J) IL-4, (K) IL-5, and (L) IL-13 for T cells co-cultured with cDC1 (yellow-orange symbols), cDC2 (red symbols), DC3 (purple symbols) and pDC (blue symbols) of six donors (each donor with an individual symbol). Statistical analysis was performed in GraphPad Prism (V10) using 2way ANOVA for grouped data with Dunnett’s multiple comparisons tests as posthoc test (*p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001).