Early Effector CD8 T Cells Display Plasticity in Populating the Short-Lived Effector and Memory-Precursor Pools Following Bacterial or Viral Infection

Abstract

Naïve antigen-specific CD8 T cells expand in response to infection and can be phenotypically separated into distinct effector populations, which include memory precursor effector cells (MPECs) and short-lived effector cells (SLECs). In the days before the peak of the T cell response, a third population called early effector cells (EECs) predominate the antigen-specific response. However, the contribution of the EEC population to the CD8 T cell differentiation program during an antimicrobial immune response is not well understood. To test if EEC populations were pre-committed to either an MPEC or SLEC fate, we purified EECs from mice infected with Listeria monocytogenes (LM) or vesicular stomatitis virus (VSV), where the relative frequency of each population is known to be different at the peak of the response. Sorted EECs transferred into uninfected hosts revealed that EECs were pre-programmed to differentiate based on early signals received from the distinct infectious environments. Surprisingly, when these same EECs were transferred early into mismatched infected hosts, the transferred EECs could be diverted from their original fate. These results delineate a model of differentiation where EECs are programmed to form MPECs or SLECs, but remain susceptible to additional inflammatory stimuli that can alter their fate.

Figure 1. EEC emerge following multiple infection early after infection.

A. OVA-specific CD8 T cells were identified as: CD8+, OVA MHC class I tetramer+, CD11a^hi within the splenocytes at the peak of the T cell response to LM-OVA (day 8), VSV-OVA (day 7), Influenza A-OVA (day 10), and Vaccinia-OVA (day 7) and analyzed for KLRG1 and CD127 expression. In the pie charts, KLRG1+ CD127− cells are classified as SLECs, KLRG1− CD127+ as MPECs, KLRG1− CD127− as EECs and KLRG1+ CD127+ as DPECs for each infection. B. 1,000 CD45.1+ transgenic OT-I T cells were transferred to uninfected CD45.2+ recipients one day prior to infection with LM-OVA or VSV-OVA. At days 5, 6 and 7 post-infection, CD45.1+ OT-I T cells were analyzed in the spleen for KLRG1 and CD127. Percentages of SLECs, MPECs, and EECs from each condition are graphed from 3–4 mice. This data is representative of at least 2 experiments.

Figure 2. EECs differentiate based on previously received inflammatory signals in uninfected recipients.

A. Schematic of EEC sorting strategy. 1,000 CD45.1+ OT-I cells were transferred to uninfected CD45.2+ recipients one day prior to infection with LM-OVA or VSV-OVA. 5 days following infection, CD8+ CD45.1+ KLRG1− CD127− EECs were sorted to ~99% purity. B. EECs were purified, as described in A, from both LM-OVA- and VSV-OVA-infected hosts and transferred to uninfected recipients. CD8+ CD45.1+ transferred EECs were then analyzed for KLRG1 and CD127 expression from splenocytes 3 or 6 days post-transfer. Percentages of SLECs, MPECs, and EECs from each condition are graphed from 3–4 mice. This data is representative of at least 2 experiments.

Figure 3. Whole genome microarray studies reveal a subset of genes differentially expressed in EECs generated after LM versus VSV infection.

A. mRNA Expression levels were measured by microarray analysis with Illumina BeadChips of sorted EECs, as in Fig. 2A, 5 days following LM-OVA or VSV-OVA infection. The top 50 and bottom 50 differentially regulated genes, ranked by signal-to-noise, are displayed. Each row represents an individual sample. B. The AVG signal from the microarray of LM or VSV EECs for selected MPEC genes is graphed. C. Enrichment profile generated from GSEA using the top 50 genes enriched in VSV EECs compared to a gene set generated with IL-7R^Hi (MPEC) and IL-7R^Lo (SLEC) samples. Genes identified as being enriched in both VSV EECs and IL-7R^Hi cells (MPECs) are marked with a red asterisk in part A.

Figure 4. EECs transferred into infected mice show plasticity by modulating their differentiation according to the infection-induced inflammatory environment of the recipient mouse.

A. EECs were purified, as described in Fig. 2A, following LM-OVA or VSV-OVA infection and transferred into LM-OVA infected recipients that had been infected for 2 days. 6 days following the EEC transfer, CD8+ CD45.1+ transferred EECs in the spleen were analyzed for KLRG1 and CD127. Percentages of SLECs, MPECs, and EECs are graphed for both LM-OVA EECs and VSV-OVA EECs from 3–4 mice. This data is representative of at least 2 experiments. B. Same as A, except recipient mice were infected with VSV-OVA 2 days prior to EEC transfer.

Figure 5. Optimal SLEC generation and proliferation requires continued antigen expression.

A. EECs were purified, as described in Fig. 2A, following VSV-OVA infection and transferred to recipient mice infected with LM-OVA, VSV-OVA, LM or left uninfected. 6 days following transfer, the CD8+, CD45.1+ EECs in the spleen were analyzed for KLRG1 and CD127 expression. Each condition was repeated at least 2 times. B. Percentages of SLECs, MPECs, and EECs are graphed for both LM-OVA EECs and LM EECs from 3–4 mice. C. Total numbers of CD8+ CD45.1+ transferred EECs per spleen, from the experiment in part A are shown graphically. D. EECs from both VSV-OVA- and LM-OVA-infected mice were sort purified as in Fig. 2A. Following purification, CD45.1+ EECs were mixed with CD45.2+ splenocytes from a naïve mouse at a 1:10 ratio prior to CFSE labeling. Total CFSE labeled cells were then transferred to uninfected recipients and CD8 CD45.1+ EECs were analyzed for KLRG1 and CD127 expression and CFSE dilution 3 days following transfer. The CFSE labeled CD45.2+ splenocytes were used as the undivided control.

Figure 6. With time, EECs lose their ability to generate MPEC and SLEC in the absence of exogenous inflammatory stimuli.

A. 14 days following VSV-OVA infection, EECs were purified as in Fig. 2A and transferred into day 2 LM-OVA- or VSV-OVA-infected recipients. 6 days following EEC transfer, CD8 CD45.1+ EECs were analyzed for KLRG1 and CD127 expression. Each condition was repeated at least 2 times. B. Day 14 purified EECs from VSV-OVA infection were transferred into uninfected recipients, and CD8 CD45.1+ EECs in the spleen were analyzed for KLRG1 and CD127 expression 3 or 6 days following transfer. Each condition was repeated at least 2 times.