A critical role for direct TLR2-MyD88 signaling in CD8 T-cell clonal expansion and memory formation following vaccinia viral infection

Abstract

Recent advances have suggested a crucial role of the innate immunity in shaping adaptive immune responses. How activation of innate immunity promotes adaptive T-cell responses to pathogens in vivo is not fully understood. It has been thought that Toll-like receptor (TLR)-mediated control of adaptive T-cell responses is mainly achieved by the engagement of TLRs on antigen-presenting cells to promote their maturation and function. In this study, we showed that direct TLR2-myeloid differentiating factor 88 (MyD88) signaling in CD8 T cells was also required for their efficient clonal expansion by promoting the survival of activated T cells on vaccinia viral infection in vivo. Effector CD8 T cells that lacked direct TLR2-MyD88 signaling did not survive the contraction phase to differentiate into long-lived memory cells. Furthermore, we observed that direct TLR2 ligation on CD8 T cells promoted CD8 T-cell proliferation and survival in vitro in a manner dependent on the phosphatidylinositol 3-kinase (PI3K)-Akt pathway activation and that activation of Akt controlled memory cell formation in vivo. These results identify a critical role for intrinsic TLR2-MyD88 signaling and PI3K-Akt pathway activation in CD8 T-cell clonal expansion and memory formation in vivo and could lead to the development of new vaccine approaches.

Figure 1

Endogenous MyD88^−/− and TLR2^−/− CD8 T cells exhibit defective virus-specific effector response to VV infection in vivo. A total of 10⁶ purified WT, MyD88^−/−, or TLR2^−/− polyclonal CD8 T cells were transferred along with 2 × 10⁶ WT polyclonal CD4 T cells into RAG-2^−/− hosts. Seven days after transfer, mice were infected intraperitoneally with 5 × 10⁶ pfu VV or left uninfected (naive) as control. Seven days after infection, splenocytes were stained with anti-CD4 and anti-CD8 antibodies and subjected to intracellular staining with an anti–IFN-γ antibody after a 6-hours stimulation with the VV-specific peptide F2L (²⁶SPYAAGYDL³⁴). (A) The percentage of IFN-γ⁺ CD8⁺ T cells among total splenocytes for each respective population is indicated. (B) The absolute cell number per spleen of IFN-γ⁺CD8⁺ T cells for both naive and infected groups are shown with SDs indicated (n = 6 per group). For all groups: naive versus VV, P < .001. For VV-infected hosts: MyD88^−/− or TLR2^−/− versus WT, P < .001. Data shown are representative of 3 independent experiments.

Figure 2

Clonal expansion is severely compromised in MyD88^−/− and TLR2^−/− clone 4 CD8 T cells in response to viral infection in vivo. A total of 10⁴ naive WT, MyD88^−/−, or TLR2^−/− clone 4 CD8 T cells (Thy1.1⁺) were transferred into WT (WT→WT, MyD88^−/− →WT, or TLR2^−/−→WT) or MyD88^−/− (WT→MyD88^−/−) recipient mice (Thy1.2⁺) that were either left uninfected (naive) or infected with 5 × 10⁵ pfu rVV-HA intraperitoneally. Seven days after infection, splenocytes were analyzed for the expansion of clonotypic T cells and their function by IFN-γ intracellular staining. (A) The percentages of clonotypic T cells among total lymphocytes (top) and IFN-γ–producing clonotypic T cells among total CD8⁺ T cells (bottom) in rVV-HA infected mice are indicated. (B) The absolute cell number per spleen of CD8⁺Thy1.1⁺ clonotypic T cells in both uninfected and rVV-HA infected hosts with SDs are indicated (n = 4 per group). For all groups: naive versus rVV-HA, P < .001. For rVV-infected hosts: WT→MyD88^−/− versus WT→WT, P < .05; MyD88^−/−→WT or TLR2^−/−→WT versus WT→WT, P < .001. (C) The mean fluorescence intensity (MFI) of IFN-γ–producing clonotypic cells is indicated. WT→MyD88^−/−, MyD88^−/−→WT, or TLR2^−/−→WT versus WT→WT, P > .05. Data shown are representative of 3 independent experiments.

Figure 3

MyD88 signaling in CD8 T cells is critical for their survival after activation and proliferation in vivo. (A,B) A total of 10⁶ WT or MyD88^−/− clone 4 CD8 T cells (Thy1.1⁺) were transferred into B10.D2 mice, which were infected with 5 × 10⁶ pfu rVV-HA intraperitoneally or left uninfected (naive). (A) At 24 hours after infection, splenocytes were analyzed for T-cell activation phenotypically by staining with anti-CD44 and anti-CD62L (left panels) or anti-CD25 and anti-CD69 antibodies (right panels). FACS plots are gated on CD8⁺Thy1.1⁺ clonotypic cells, and percentages shown are of those cells possessing an activated phenotype of CD44^highCD62L^low or CD25^highCD69^high. (B) Three days after infection, proliferation of the clonotypic cells was analyzed by the CFSE dilution assay. Events were gated on CD8⁺Thy1.1⁺ cells. (C) A total of 10⁴ WT or MyD88^−/− clone 4 CD8 T cells were transferred into B10.D2 mice that were either left uninfected (naive) or infected with 5 × 10⁵ pfu rVV-HA (rVV-HA). Seven days after infection, splenocytes were harvested and the extent of clonotypic cells undergoing apoptosis was analyzed by annexin V staining. Events were gated on CD8⁺Thy1.1⁺ cells, and percentages shown are of those cells that are annexin V⁺. Representative data of 3 independent experiments with 3 mice per group for each experiment are shown.

Figure 4

A critical role for direct TLR2-MyD88 signaling in the formation of memory CD8 T cells. A total of 10⁴ naive WT, MyD88^−/−, or TLR2^−/− clone 4 CD8 T cells (Thy1.1⁺) were transferred into WT (WT→WT, MyD88^−/− →WT, or TLR2^−/−→WT) or MyD88^−/− (WT→MyD88^−/−) recipient mice (Thy1.2⁺) that were infected with 5 × 10⁵ pfu rVV-HA intraperitoneally. (A) Splenocytes were harvested at days 0, 7, 14, 28, or 42 after infection to determine the absolute number of the CD8⁺Thy1.1⁺ population per spleen, shown with SDs indicated (n = 4 per group). (B) At 42 days after infection, splenocytes were stained with anti-CD8 and anti-Thy1.1. The percentages of clonotypic T cells among total lymphocytes (top) and IFN-γ–producing clonotypic T cells among total CD8⁺ T cells (bottom) are indicated. Data shown are representative of 3 independent experiments.

Figure 5

Intrinsic TLR2-MyD88 signaling increases cellular proliferation and survival after in vitro stimulation and is dependent on the PI3K pathway. (A,B) Polyclonal CFSE-labeled WT or MyD88^−/− CD8 T cells were stimulated in vitro with plate-bound anti-CD3 antibody alone (αCD3) or anti-CD3 antibody coupled with plate-bound anti-CD28 (αCD3 αCD28), Pam₃Cys (αCD3 Pam₃Cys), Pam₃Cys and the PI3K inhibitor LY294002 (αCD3 Pam₃Cys LY294002), or left unstimulated (naive) as a control. After 18 hours of stimulation, T cells were removed from stimulation and placed back into culture in the absence of further stimulation for a total of 4 days. At this time, both the CFSE profile (A) as well as the survival of CD8⁺ cells by annexin V staining (B) were determined by flow cytometry. Percentage of annexin V⁻ cells among total CD8⁺ T cells is indicated. Representative data from 4 independent experiments are shown.

Figure 6

Addition of Pam₃Cys increases activation of the PI3K-Akt pathway in polyclonal CD8 T cells. Polyclonal CD8 T cells were stimulated in vitro with plate-bound anti-CD3 antibody alone (αCD3 only), coupled with either Pam₃Cys (αCD3 + Pam₃Cys) or plate-bound anti-CD28 (αCD3 + αCD28), or left unstimulated (naive) as a control. After 6, 12, and 18 hours of stimulation, CD8 T cells were removed from culture and total cell lysates were collected for Western blot analysis of phosphorylated (pAkt) as well as total Akt, which served as a loading control. Data shown are a representative blot of 5 independent experiments.

Figure 7

Activation of and signaling through Akt is required for CD8 memory cell formation in vivo. (A,B) Naive WT clone 4 CD8⁺ T cells were transduced with retroviral constructs encoding GFP only (empty) as a control or GFP coupled with a dominant-negative form of Akt (dnAkt). The transduced cells were sorted, and 3 × 10³ GFP⁺CD8⁺ (Thy1.1⁺) T cells were transferred into B10.D2 recipients (Thy1.2⁺), which were then vaccinated with 5 × 10⁵ pfu rVV-HA intraperitoneally. At 42 days later, mice were boosted with 2 × 10⁹ pfu Ad-HA intraperitoneally. Seven days after infection, splenocytes were analyzed for the expansion of clonotypic T cells and their function by IFN-γ intracellular staining. (A) The percentages of clonotypic T cells among total lymphocytes (top) and IFN-γ–producing clonotypic T cells among total CD8⁺ T cells (bottom) are indicated. (B) The absolute cell numbers per spleen of CD8⁺Thy1.1⁺ clonotypic T cells in both empty vector and dnAkt groups with SDs are indicated (n = 4 per group). Empty versus dnAkt, P < .001. Data shown are representative of 3 independent experiments.