Kruppel-like factor 2 is required for trafficking but not quiescence in postactivated T cells

Abstract

The transcription factor Kruppel-like factor 2 (KLF2) was proposed to regulate genes involved in cell cycle entry and T cell trafficking; however, the physiological role of its expression in postactivated T cells is not well defined. Previous studies suggested that the cytokines IL-2 and IL-15 differentially regulate KLF2 re-expression in postactivation T cells and that these cytokines also influence effector versus memory T cell differentiation. Using conditional and inducible KLF2-knockout model systems, we tested the specific role of KLF2 expression in activated CD8(+) T cells cultured with these cytokines. KLF2 was required for effective transcription of sphingosine-1-phosphate receptor-1 (S1P(1)) and CD62L in postactivation T cells. However, although different cytokines dramatically altered the expression of cell-cycle-related genes, endogenous KLF2 had a minimal impact. Correspondingly, KLF2-deficient T cells showed dysregulated trafficking but not altered proliferative characteristics following in vivo responses to Ag. Thus, our data help to define KLF2-dependent and -independent aspects of activated CD8(+) T cell differentiation and argue against a physiological role in cell cycle regulation.

Figure 1. Differential cytokine effects on post-activated CD8+ T cells

Spleen cells from Rag−/−OT-I mice were stimulated with OVAp for 2 days. Viable cells were subsequently cultured in the presence of IL-2, IL-7 or IL-15 for another 6 days. (A) RNA samples were obtained at day 2 or 8 of the stimulation culture and subjected to real time PCR analysis of indicated genes. (B) Expression of CD62L and CD69 on CD8+ T cells at the end of culture (day 8). Representative of 2–4 independent experiments with similar results.

Figure 2. KLF2-deficient cells are functionally mature

(A) Phenotype of naïve CD4SP thymocytes from KLF2^fl/flCD4Cre (KO) or control (WT) mice. (B and C). KLF2^fl/flCD4Cre (KO) or control (WT) thymocytes were stimulated with anti-CD3 and anti-CD28 in vitro. In (B) the dot plots shows Qa2 expression and Annexin V binding (18 hours after stimulation) on CD4SP cells. The bar graph shows quantification of specific apoptosis in Qa2hi versus Qa2lo CD4SP cells. In (C) Thymocytes were labeled with CFSE prior to stimulation. The dot plots show Qa2 expression and CFSE (48 hours after stimulation) on CD4SP gate. Histograms show CFSE dilution of Qa2hiCD4SP cells cultured with or without anti-CD3 and anti-CD28. (D) Thy1.1+CD8+CD4− thymocytes from OT-I KLF2−/+ (WT) or KLF2−/− (KO) fetal liver chimeras were transferred to B6 mice (0.75 × 10⁵ cells/mouse). Recipient mice were injected i.p. with 5 × 10⁶ PFU of vaccinia virus expressing OVA. Graphs show the number and frequency (in the live cell gate) of donor cells in different organs 5 days after infection. Representative of 2–3 independent experiments with similar results.

Figure 3. KLF2 deficiency does not affect the quiescence of post-activated CD8+ T cells

Thymocytes were obtained from WT (KLF2^fl/flCD4Cre) and KO (KLF2^fl/fl) mice and were stimulated with anti-CD3 and irradiated splenocytes for 2 days. Viable cells were cultured in the presence of IL-2 or IL-15 for additional 6 days. (A) Shows changes of CD8+ T cell phenotype at various time points after activation. (B) At the end of culture (day 8), CD8+ cells were enriched by magnetic beads, and RNA was extracted. Expression of indicated genes were evaluated by real time PCR. Values are relative to the level of IL-2-treated WT group which was defined 1.0. (C) At day 8, cells were labeled with CFSE and further cultured with IL-15 for additional 24 and 72 hours. Histograms show the CFSE dilution in CD8SP gate. (D) At day 8, viable cells were enriched and Annexin V staining was examined before (0 hour) and after the 24-hour-culture in the presence of IL-15. Apoptosis during these 24 hours was calculated as described in materials and methods. Representative of 2–4 independent experiments with similar results.

Figure 4. Inducible loss of KLF2 after activation impairs the expression of trafficking- but not quiescence-associated molecules

Spleen cells from KLF2^fl/fl (F/F) or KLF2^+/+ (+/+) mice bearing ROSA26-floxSTOP-YFP allele were stimulated with plate-bound anti-CD3 for 2 days, then cultured with Tat-Cre for 45 minutes. Viable cells were cultured in the presence of IL-2 or IL-15 for additional 6 days. (A) Phenotype of CD8+YFP+ cells at the end of the culture. (B) CD8+YFP+CD44hi cells were sorted from the IL-15-treated group. RNA was extracted and used for the template of the real-time PCR analysis for the indicated genes. Expression levels are shown relative to WT group defined as 1.0. (C) IL-15-treated F/F or +/+ cells (CD45.2+Thy1.2+) were mixed with equal number (5 × 10⁶) of naïve splenocytes from B6. PL mice (CD45.2+Thy1.1+) and injected intravenously into B6. SJL mice (CD45.1+Thy1.2+). Twenty-four hours after adoptive transfer, the blood, spleen (SPL), peripheral lymph nodes (PLN), mesenteric lymph nodes (MLN) from recipients were analyzed for the frequency of CD45.2+Thy1.1-YFP+CD8+ cells (test cells) and CD45.2+Thy1.1+CD8+ cells (internal control). Homing index was calculated as described in materials and methods. Representative of 2–3 independent experiments with similar results.

Figure 5. KLF2-deficient T cells respond normally to infection but show aberrant distribution

CD8SP thymocytes from KLF2^fl/flCD4Cre OT-I Tg (KO) or control OT-I Tg (WT) mice were mixed at equal number and co-transferred to B6 mice (2 × 10⁴ cells from each group/recipient). One day later, recipient mice were injected i.p. with 3 × 10⁶ CFU of attenuated L. monocytogenes expressing OVA. (A) Changes of donor CD8+ T cell numbers over time following infection in the indicated tissues. Total number indicates the sum of donor cell counts in the spleen, lymph nodes, and the liver. (B) Comparison of the KO and WT cell numbers (left graph) in the spleen (SPL), lymph nodes (LN) and the liver (LIV) or the percentages in the live cell gate in the blood (right graph) at day 30 after infection. (C) Histograms show the expression of CD62L and CD69 on the surface of donor CD8+ T cells at day 30 after infection. Representative of 3 independent experiments with similar results.