Reciprocal and dynamic control of CD8 T cell homing by dendritic cells from skin- and gut-associated lymphoid tissues

Abstract

T cell activation by intestinal dendritic cells (DC) induces gut-tropism. We show that, reciprocally, DC from peripheral lymph nodes (PLN-DC) induce homing receptors promoting CD8 T cell accumulation in inflamed skin, particularly ligands for P- and E-selectin. Differential imprinting of tissue-tropism was independent of Th1/Th2 cytokines and not restricted to particular DC subsets. Fixed PLN-DC retained the capacity to induce selectin ligands on T cells, which was suppressed by addition of live intestinal DC. By contrast, fixed intestinal DC failed to promote gut-tropism and instead induced skin-homing receptors. Moreover, the induction of selectin ligands driven by antigen-pulsed PLN-DC could be suppressed "in trans" by adding live intestinal DC, but PLN-DC did not suppress gut-homing receptors induced by intestinal DC. Reactivation of tissue-committed memory cells modified their tissue-tropism according to the last activating DC's origin. Thus, CD8 T cells activated by DC acquire selectin ligands by default unless they encounter fixation-sensitive signal(s) for gut-tropism from intestinal DC. Memory T cells remain responsive to these signals, allowing for dynamic migratory reprogramming by skin- and gut-associated DC.

Figure 1.

Reciprocal control of T cell homing potential by DC from skin-draining LN and PP. Naïve P14 CD8 T cells were cultured with peptide-pulsed PLN-DC or PP-DC and analyzed after different time intervals for the expression of skin- (E- and P-selectin ligands) and gut-homing molecules (α4β7 and CCR9). (A) Kinetics of the induction of skin- and gut-homing molecules on CD8 T cells activated with PLN-DC or PP-DC. (B) Representative experiment showing the expression of skin- or gut-homing molecules on CFSE-labeled CD8 T cells activated with PLN-DC or PP-DC after 3 d of coculture. Vertical quadrant markers were set so that undivided CFSE-labeled input cells fall in the right quadrants. Arrows and numbers indicate cell divisions. Percent of gated cells in upper left quadrants is also shown. Relationship between the frequency (C) and MFI (D) of T cells expressing skin- and gut-homing molecules and the number of cells divisions of CD8 T cells activated with PLN-DC or PP-DC after 3 d of coculture. Gates were set on viable CD8⁺ T cells. Statistical analyses were performed using a paired Student's t test comparing CD8 T cells activated with PLN-DC vs. those activated with PP-DC (mean ± SEM, n = 5–16).

Figure 2.

mRNA expression, chemotactic responsiveness, and integrin levels of CD8 T cells activated with PLN-DC or PP-DC. After different time intervals, CD8 T cells activated with PLN-DC or PP-DC were analyzed by real-time PCR for their relative mRNA expression of (A) fucosyltransferase-VII (FucT-VII), (B) the skin-associated chemokine receptor CCR4, as well as the gut homing molecules α4 integrin chain (C) and CCR9 (D). All mRNA levels were normalized with respect to GAPDH. Statistical analyses were performed using a paired Student's t test comparing CD8 T cells activated with PLN-DC vs. those activated with PP-DC (mean ± SEM, n = 3). (E) Chemotactic response toward the CCR4-ligand CCL22. Results are expressed as the percentage of total cells migrating toward a chemokine or toward medium alone. For chemotaxis experiments, cells were used after 4–6 d of coculture (mean ± SEM, n = 11). (F) Flow cytometry histograms showing expression of α4β7 heterodimer, α4, β7, αE, and β1 integrin chains on CD8 T cells activated with PLN-DC or PP-DC after 6 d of culture. Results show one representative experiment out of five performed with similar results. The marker (M1; indicating specific labeling) was set according to appropriate isotype controls.

Figure 3.

CD8 T cells activated with PLN-DC migrate more efficiently into inflamed skin and interact significantly better with cutaneous venules than those activated with PP-DC. CD8 T cells activated with PLN-DC or PP-DC were used at day 4–5 for homing experiments and IVM in mouse ear venules. (A) Competitive homing experiments between CD8 T cells activated with PLN-DC or PP-DC. Both populations were differentially labeled, mixed at a 1:1 ratio, and injected into naïve recipient mice. 18 h later, tissues were collected and the HI was calculated for each organ. IVM of T cell interactions in ear skin venules was used to determine (B) the rolling fraction and (C) the sticking fraction (n = venules/mice). (D) A cumulative rolling velocity histogram was generated by measuring rolling velocities (V_roll) of individual CD8 T cells activated with PLN-DC (median V_roll = 19 μm/s) and CD8 T cells activated with PP-DC (median V_roll = 38 μm/s) in ear venules by plotting the percentage of all cells that rolled at or below a given V_roll (n = cells/venules/mice). (E) Correlation between V_roll and the dwell time of individual rolling cells in each venule. The dwell time was measured as the time a T cell spent rolling within the venular segment that was used to measure V_roll. To compare homing experiments, results were analyzed using a one-sample t test (comparing experimental HI vs. HI = 1). To compare rolling and sticking fractions, a paired Student's t test was used comparing CD8 T cells activated with PLN-DC vs. those activated with PP-DC. The Mann-Whitney U-test was employed to compare V_roll of CD8 T cells activated with PLN-DC vs. those activated with PP-DC, and the Spearman rank correlation test was used for analysis of V_roll vs. dwell time.

Figure 4.

Effect of sorted DC subsets on the acquisition of selectin ligands and α4β7 on CD8 T cells. (A) DC from skin-draining LN or PP were isolated by negative selection and then subdivided by FACS sorting into CD11c^LoB220⁺ (‘plasmacytoid’ DC) and CD11c^HiB220^Neg DC. The latter were further sorted into CD8α^NegCD11b^Hi (“myeloid” DC), CD8α^HiCD11b^Lo (“lymphoid” DC), and CD8α^Lo/-CD11b^Lo subsets. Each DC subpopulation was pulsed with peptide Ag and cocultured with naïve CD8 T cells for 5–6 d. Expression of (B) E- and (C) P-selectin ligands, (D) α4β7, and (E) LFA-1 on CD8 T cells activated with DC subsets from PLN or PP. Gates were set on viable CD8⁺ T cells. Statistical analyses were performed using a paired Student's t test comparing CD8 T cells activated with each PLN-DC subset vs. those activated with its PP-DC counterpart (mean ± SEM; n = 2–4).

Figure 5.

Effect of Th1/Th2 cytokines on the expression of selectin ligands, α4β7, and CCR9 on CD8 T cells activated with PLN-DC and PP-DC. (A) Naïve CD8 T cells were cultured with peptide-pulsed PLN-DC or PP-DC, either in the presence of blocking antibodies against IL-12, IFN-γ, or IL-4, and/or the cytokines themselves. CD8 T cells were analyzed after 4–5 d of coculture for expression of (A) E- and (B) P-selectin ligands, (C) α4β7, and (D) CCR9. Cells were gated on viable (FSC/SSC) CD8⁺ T cells. Results were normalized in each experiment to values from a parallel culture of untreated CD8 T cells activated with PP-DC (assigned a value of 100). To compare CD8 T cells activated with PP-DC or PLN-DC in the presence of cytokines and/or mAbs vs. the respective untreated CD8 T cells activated with the same DC, an ANOVA with Dunnett's post-test was used (*, P < 0.05; ***, P < 0.001). To compare CD8 T cells activated with PLN-DC vs. CD8 T cells activated with PP-DC in each group, a two-tailed paired Student's t test was used (#, P < 0.05; ###, P < 0.001). Only groups with n ≥ 3 independent experiments were used for statistical comparison. Bars reflect mean ± SEM.

Figure 6.

Effect of DC fixation and mixed DC cultures on homing molecule expression by T cells. (A–D) Naïve CD8 T cells were cultured with peptide-pulsed PLN-DC or PP-DC, which were either left untreated or fixed with 0.05% glutaraldehyde. CD8 T cells were analyzed after 4 d of coculture for expression of (A) E- and (B) P-selectin ligands, (C) α4β7, and (D) CCR9. Cells were gated on viable (FSC/SSC) CD8⁺ T cells. Results in A–D are expressed as percentage of marker-positive CD8 T cells. Bars represent mean ± SEM (mean ± range when n < 3). To compare CD8 T cells activated with unfixed PP-DC alone to cells activated under any other condition, an ANOVA with Dunnett's post-test was employed. (E–H) Naïve CD8 T cells (10⁵/well) were cocultured with an equivalent number of Ag-pulsed PLN-DC (E and G) or PP-DC (F and H), with or without addition of unpulsed PLN-DC or PP-DC. Cells were analyzed after 4 d for the expression of selectin ligands (E and F) or α4β7 heterodimer and CCR9 (G and H). Cells were gated on viable CD8⁺ T cells. Results were normalized in each experiment to values from parallel cultures of CD8 T cells with Ag-pulsed PLN-DC (E and G) or PP-DC (F and H), which were arbitrarily assigned a value of 100. Symbols represent the average of two experiments with comparable results.

Figure 7.

Preservation of plasticity of ex vivo–generated tissue-tropic effector CD8 T cells. Naïve CD8 T cells were activated with either PLN-DC or PP-DC (1^st activation) and the resulting effector T cells were restimulated 4–5 d later with either PLN-DC or PP-DC (2^nd activation). 4–5 d after the second activation, effector cells were compared with naïve T cells (N) for expression of ligands for (A) E-selectin and (B) P-selectin or (C) α4β7 and (D) CCR9. Statistically significant (P < 0.05) differences were found for all markers analyzed between CD8 T cells activated with PLN-DC vs. those activated with PP-DC irrespective of the starting population. Statistical analyses were performed using ANOVA with Newman-Keuls post-test, comparing all groups of effector CD8 T cells at day 8–9 (mean ± SEM, n = 3-4).

Figure 8.

Plasticity of endogenous tissue-tropic effector–memory CD8 T cells. CD8 T cells were isolated by negative selection from pooled cells from spleen, PLN, MLN, and PP from 4–6-mo-old C57BL/6 wild-type mice, and the CD8⁺CD44^High fraction was sorted by FACS into (A) P-Lig⁺ and P-Lig^Neg subsets or (B) CCR9⁺ and CCR9^Neg subsets. The sorted cells were plated on immobilized anti-CD3+anti-CD28 and supplemented with either PLN-DC or PP-DC. The resulting effector CD8 T cells were analyzed 4 d later for their capacity to bind E- and P-selectin and their expression of α4β7 and CCR9. Statistical analyses were performed using a paired Student's t test comparing effector–memory CD8 T cells activated with PLN-DC vs. those activated with PP-DC (mean ± SEM, n = 3).