Desirable cytolytic immune effector cell recruitment by interleukin-15 dendritic cells

Abstract

Success of dendritic cell (DC) therapy in treating malignancies is depending on the DC capacity to attract immune effector cells, considering their reciprocal crosstalk is partially regulated by cell-contact-dependent mechanisms. Although critical for therapeutic efficacy, immune cell recruitment is a largely overlooked aspect regarding optimization of DC vaccination. In this paper we have made a head-to-head comparison of interleukin (IL)-15-cultured DCs and conventional IL-4-cultured DCs with regard to their proficiency in the recruitment of (innate) immune effector cells. Here, we demonstrate that IL-4 DCs are suboptimal in attracting effector lymphocytes, while IL15 DCs provide a favorable chemokine milieu for recruiting CD8+ T cells, natural killer (NK) cells and gamma delta (γδ) T cells. Gene expression analysis revealed that IL-15 DCs exhibit a high expression of chemokines involved in antitumor immune effector cell attraction, while IL-4 DCs display a more immunoregulatory profile characterized by the expression of Th2 and regulatory T cell-attracting chemokines. This is confirmed by functional data indicating an enhanced recruitment of granzyme B+ effector lymphocytes by IL-15 DCs, as compared to IL-4 DCs, and subsequent superior killing of tumor cells by the migrated lymphocytes. Elevated CCL4 gene expression in IL-15 DCs and lowered CCR5 expression on both migrated γδ T cells and NK cells, led to validation of increased CCL4 secretion by IL15 DCs. Moreover, neutralization of CCR5 prior to migration resulted in an important inhibition of γδ T cell and NK cell recruitment by IL-15 DCs. These findings further underscore the strong immunotherapeutic potential of IL-15 DCs.

Keywords: CCL4-CCR5 signaling; NK cells; dendritic cell vaccination; immune cell recruitment; γδ T cells.

Figure 1. Immune effector cells are recruited by IL-15 DCs

48-hour wash-out supernatant of IL-15 DCs and IL-4 DCs was used in a three-hour 5 μm pore size transwell chemotaxis assay with PBMC. Scatter dot plots (line at mean) represent the % increase of migration, calculated as (number of migrated cells in the specific condition / number of migrated cells in the negative control) × 100, of viable migrated PBMC subsets (n = 6), with the negative control representing spontaneous migration towards medium. Immune cells recruited by IL-15 DC and IL-4 DC wash-out supernatant are depicted as dark grey and light grey spheres, respectively. Repeated measures ANOVA with Bonferroni's Multiple Comparison Testing was used for statistical analysis. Statistics depicted on top of the brackets represent differences between migration towards supernatant of IL-15 DCs versus IL-4 DCs, whereas statistics of migration towards each DC type as compared to the medium control is noted below the brackets. ***p < 0.001, **p < 0.01, *p < 0.05, ns, p > 0.05

Figure 2. DC-mediated migration and phenotype analysis of purified γδ T cells

(A) Percentage of isolated γδ T cell migration towards medium (control), 48-hour wash-out supernatant of IL-15 DCs (IL-15 DCs) and IL-4 DCs (IL-4 DCs). Data were calculated as percentage of migrated γδ T cells relative to background migration and show mean values (+ SD) of eleven different donors. (B) Bar graphs illustrate the relative distribution of Live/dead− CD3+ γδ TCR+ TCR-δ2+ and Live/dead- CD3+ γδ TCR+ TCR-δ1+ subsets of purified γδ T cells (γδ T cell) and following DC-mediated migration (n = 7). (C) Percentage surface expression of chemokine receptors CCR2, CCR5, CCR7 and CXCR3, and IL-15Rα on isolated Live/dead- CD3+ γδ TCR+ T cells before and after migration (n = 6–10). For δ1/δ2 distribution and CCR5/CCR7 expression) a repeated measures ANOVA with Bonferroni's Multiple Comparison Test was used, for γδ T cell migration, CCR2/CXCR3/IL-15Rα expression a Friedmann Test with Dunn's Multiple Comparison Test was used. ***p < 0.001, *p < 0.05.

Figure 3. DC-mediated migration and phenotype analysis of purified NK cells

(A) Migration capacity of purified NK cells towards the chemokine milieu of IL-15 DCs and IL-4 DCs is depicted as % increase migration (+ SD) as compared to the negative control (n = 11). (B) Bar graphs represent the relative distribution of Live/dead- CD3- CD56^bright and Live/dead- CD3− CD56^dim subsets in isolated NK cells prior to migration (NK cell) and in NK cells migrated towards IL-15 DC (IL-15 DCs) and IL-4 DC (IL-4 DCs) wash-out supernatant. (C) Percentage surface expression of chemokine receptors CCR2, CCR5, CCR7 and CXCR3, and IL-15Rα on isolated Live/dead− CD3− CD56+ NK cells before and after migration (n = 8). Repeated measures ANOVA with Bonferroni's Multiple Comparison Test (NK cell migration, IL-15Rα expression), Friedmann Test with Dunn's Multiple Comparison Test (CD56 distribution, chemokine receptor expression). ***p < 0.001, **p < 0.01, *p < 0.05.

Figure 4. Hierarchical cluster analysis of 18 genes belonging to the chemokine family, distinctly different between IL-15 DCs and IL-4 DCs

The heat map represents the differential expressed genes patterns between IL-15 DC samples (left) and IL-4 DC samples (right) of three independent donors. The value range represents the Log2 of the ratio between expression signals and the global median expression for each gene. Red color indicates up-regulated genes and green color down-regulated genes.

Figure 5. CCR5-dependent recruitment of effector immune cells by IL-15 DCs

(A) Quantitative comparison of CCL4 secretion by IL-15 DCs and IL-4 DCs as measured by ELISA in 48-hour wash-out supernatant of IL-15 DCs and IL-4 DCs. Concentration (pg/mL + SD) is shown of duplicate conditions for 10 donors. Paired T test. (B) Time course of CCL4 secretion (pg/mL) by IL-15 DCs (dark grey lines) and IL-4 DCs (light grey lines) determined by ELISA in 4-, 12-, 24- and 48-hour wash-out supernatant (in duplicate) of 4 different donors (unique symbols). (C) PBMC were cultured in the absence (-) or presence of neutralizing anti-CCR5 mAbs (αCCR5) for 1 hour prior to a three-hour chemotaxis assay towards medium (control), 48-hour wash-out supernatant of IL-15 DCs (IL-15 DCs) or IL-4 DCs (IL-4 DCs). Data of 6 independent donors (unique symbols) were calculated as percentage of migrated γδ T cells and NK cells normalized to the negative control. Wilcoxon matched-pairs signed rank test. Different donors were used for experiment B and C. **p < 0.01, *p < 0.05

Figure 6

IL-15 DCs superiorly attract cytotoxic immune effector cells (A). Floating bars (min to max, line at mean) represent % granzyme B+ cells in PBMC (white bar), PBMC exposed for 3 hours to 48-hour wash-out supernatant of IL-15 DCs (striped dark grey bar) or IL-4 DCs (striped light grey bar), and PBMC following IL-15 DC- (dark grey bar) or IL-4 DC-mediated (light grey bar) migration (n = 3). (B) Representation of % granzyme B+ cells (+ SD) within the different effector cell subsets in full PBMC fraction (white bars) and following IL15 DC- (dark grey bars) or IL-4 DC-mediated (light grey bars) migration (n = 8). (C) Cytotoxicity was determined of migrated purified γδ T cells and NK cells towards 48-hour wash-out supernatant of IL-15 DCs (IL-15 DCs) or IL4 DCs (IL-4 DCs). Non-migrated purified cells (-) were used to define control killing capacity. Daudi and K562 target cells were added at an E:T ratio of 5:1 γδ T cells and 1:1 NK cells, respectively. Percentage tumor cell killing was determined by Annexin-V/PI staining after 4 hours and calculated using the formula specified in “Materials and methods”. Donors are represented by unique symbols (n = 6). Repeated measures ANOVA with Bonferroni›s Multiple Comparison Test. ***p < 0.001, **p < 0.01, *p < 0.05, ns, p > 0.5.