The cell cycle time of CD8+ T cells responding in vivo is controlled by the type of antigenic stimulus

Abstract

A hallmark of cells comprising the mammalian adaptive immune system is the requirement for these rare naïve T (and B) lymphocytes directed to a specific microorganism to undergo proliferative expansion upon first encounter with this antigen. In the case of naïve CD8(+) T cells the ability of these rare quiescent lymphocytes to rapidly activate and expand into effector T cells in numbers sufficient to control viral and certain bacterial infections can be essential for survival. In this report we examined the activation, cell cycle time and initial proliferative response of naïve murine CD8(+) T cells responding in vivo to Influenza and Vaccinia virus infection or vaccination with viral antigens. Remarkably, we observed that CD8(+) T cells could divide and proliferate with an initial cell division time of as short as 2 hours. The initial cell cycle time of responding CD8(+) T cells is not fixed but is controlled by the antigenic stimulus provided by the APC in vivo. Initial cell cycle time influences the rate of T cell expansion and the numbers of effector T cells subsequently accumulating at the site of infection. The T cell cycle time varies with duration of the G(1) phase of the cell cycle. The duration of G(1) is inversely correlated with the phosphorylation state of the retinoblastoma (Rb) protein in the responding T cells. The implication of these findings for the development of adaptive immune responses and the regulation of cell cycle in higher eukaryotic cells is discussed.

Figure 1. Tempo of early CD8⁺ T cell division in the DLN following influenza infection.

A. In vivo division time of TCR tg CL-4 T cells responding in the DLN to influenza A/PR8 at the indicated times p.i. as determined by CFSE dye dilution. Results are representative of 6 independent experiments using 2–3 mice/exp. with mean division numbers between 72 hrs and 78 hrs pi of 3.19 ± 0.17 divisions over the 6 hr time frame. Percentages of cells at each division (#1 through 4) at 78 hr time point are 13.1% (division #1), 11.7% (#2), 8.5% (#3), and 3.5% (#4) of total CL-4 CD8⁺ T cells per DLN, respectively. B. In vivo division time of TCR tg Demi-4 T cells in the DLN responding to influenza A/JAPAN and analyzed as described above at the indicated times p.i. Results are representative of 3 individual mice analyzed in 2 experiments with a mean of 3.0 ± 0.1 divisions between 54 hrs and 60 hrs p.i. C. In vivo division time of a single cohort of CFSE labeled TCR tg CL-4 T cells responding in the DLN to infection after administration of anti CD62L mAb at 24 hrs p.i. Results are representative of 6 independent experiments with pooled DLN from 4 mice per experiment. Inserts show CFSE intensity (X-axis) and cell numbers as % of maximum (Y-axis) at the indicated time-points.

Figure 2. DNA content and DNA synthesis by responding CD8⁺ T cells in the DLN.

A. Adoptively transferred CFSE labeled CL-4 T cells were harvested from the DLN of infected mice at 78 hrs p.i. Responding cells were analyzed by flow cytometry and gated as to division number (0 through 5) based on CFSE dye intensity. DNA content was assessed by uptake of the cell permeable fluorescent dye DRAQ-5, a highly cell permeable DNA-interactive agent. Upper right panel shows the gating strategy used to identify and analyze individual dividing cell populations. Upper left panel shows DNA content of naive CL-4 T cells (in G₀/G₁) isolated from the NDLN of infected mice. Numbers shown are the percentages of cells in S+G2/M. Results are representative of the compiled experiments described in Table S1. B. At the indicated times following the i.v. administration of the thymidine analog BrdU at 78 hrs p.i., the DLN of infected recipients of CFSE dye labeled CL-4 T cells were harvested and the T cells were analyzed for CFSE intensity (cell division) and BrdU uptake (DNA synthesis activity). Data are representative of 4 independent experiments with 2–3 mice/exp. Values represent the percentage of BrdU⁺ cells among the dividing cell population. C, Results as in Figure 2A except that the DLN were harvested at days 3.5 and 5 p.i., respectively. This comparative analysis is representative of 2 independent experiments with 3 mice/exp. Numbers shown are the percentages of cells in S+G₂/M (> 2N DNA content) and the percentages of cells in G₁ (2N DNA content) for day 5 p.i. cells and the percentages of cells in S+G₂/M (> 2N DNA content) for day 3.5 p.i. cells.

Figure 3. Cell division rate, DNA content and cell cycle phase.

A. Pooled lymph nodes and spleen were harvested at the indicated times from recipients of dye labeled CL-4 T cells after i.v. administration of 100–250 µl of 1 µM soln. of synthetic PHA_533–541 peptide and the responding T cells analyzed as described in Figure 1. Results are representative of 10 independent experiments (n = 1–2 animals/experiment). Mean values for the division numbers from pooled lymph nodes and spleens are 0.66 ± 0.08 divisions from 0 hrs to 24 hrs post peptide administration, 1.28 ± 0.03 divisions for the 6 hr interval from 24 hrs to 30 hrs, and 0.99 ± 0.01 divisions between 30 hrs and 36 hrs after peptide administration. B. DNA content of CL-4 TCR tg T cells isolated either from the lymph nodes of T cell recipient mice 36 hrs after i.v. administration of synthetic peptide (PHA_533–541) or from the DLN of mice infected with influenza A/PR/8 and harvested 3.5 days (84 hr) p.i. For each group the cells are gated as divided or undivided cells and the fraction of cells with 2N or > 2N DNA content determined. Data are representative of 4 independent experiments with 3–4 mice/exp. Numbers shown are the percentages of cells in G₁ or S+G₂/M (shown are values for LN cells only). C. Percentage of dividing CL-4 T cells in the G1 phase of the cell cycle for peptide immunized or infected cell donors analyzed 3.5 days p.i. as described in Figure 3B. Values are the mean percentages from peptide treated animals (n = 13) or infected animals (n = 9) pooled from 4 independent experiments.

Figure 4. Effect of antigenic stimulus on the initial division rate of responding T cells.

A. Recipients of CFSE labeled CL-4 T cells were infected i.n. with 10⁷ pfu of recombinant vaccinia virus expressing the HA of A/PR/8. At the indicated times post infection the DLN were harvested and CL-4 T cell proliferation assessed. Results are representative of 4 independent experiments (n = 2–3 mice per experiment) with mean number of divisions of 1.55 ± 0.25 between 42 and 48 hrs p.i. and 1.0 ± 0.2 divisions between 48 hrs and 54 hrs p.i. B. Experimental design as in Figure 3A except that the PHA_533–541 peptide was delivered i.n. in a cocktail of TLR-3, 7 and 9 ligand agonists adjuvants as described (Materials and Methods). Data are representative of 2 independent experiments using 3 animals/exp. C and D. Influenza A/PR/8 infected splenic DC (C) or infectious virus (D) was delivered i.v. to recipients of CFSE labeled CL-4 T cells. At the indicated times post administration of DC or virus, lymph nodes and spleen were harvested and the proliferative response of the T cells determined. Results are representative of 2 independent experiments and a total of 6 animals. The mean number of CL-4 T cell divisions was 1.97 ± 0.2 between 30 hrs and 36 hrs for i.v. infectious virus administration.

Figure 5. Cell cycle time and T cell expansion.

A. CL-4 T cells were transferred into Thy1 congenic naïve recipients. Following i.n. infection with VV-HA or A/PR/8 virus, the DLN were harvested either immediately prior to the onset of T cell proliferation (36 hrs for VV-HA and 72 hrs 4 A/PR8) or 24 hrs thereafter. Total CL-4 T cells in the pooled DLN (2–3 donors per experiment) were enumerated at the two indicated time points. Values are the mean cell numbers from 4 experiments. B. Total CL-4 T cells accumulating in the VV-HA or Influenza infected lungs were enumerated on the indicated days. Values are pooled results from 3 independent experiments using 3–4 mice/exp.

Figure 6. Expression array profile of proliferating T cells.

CFSE labeled CL-4 T cells undergoing divisions 1 through 4 were isolated by cell sorting from the DLN of influenza infected mice (at 78 hrs p.i.) and from the LN of peptide immunized mice (at 36 hrs after peptide administration). RNA was extracted from the sorted cells (> 95% purity) and subjected to expression array analysis. Values are signal intensities for selected genes differing by 3 or more fold between the 2 comparison groups. The DLN of 8 mice/group served as the source of pooled CL-4 T cell RNA for analysis.

Figure 7. Retinoblastoma protein phosphorylation in proliferating CD8⁺ T cells.

A. Level of Rb phosphorylated at Ser_800/804 in the CFSE labeled CL-4 T cells undergoing divisions 1 through 4 isolated from the DLN of influenza infected (78 hrs p.i., solid line –––) or VV-HA infected (54 hrs p.i., dotted line··········) animals. Isotype control antibody staining intensity for CL-4 T cells from influenza infected (shaded) and VV-HA Infected (stippled) donors are included. Data are representative of 3 exp. using 3 mice/exp. B. Analysis is as in Figure 7A except that CL-4 T cells from the lymph nodes of animals immunized i.v. with the synthetic PHA_533–541 peptide epitope (36 hrs post-administration, dotted line··········) are compared to T cells from the DLN of influenza infected mice (78 hrs p.i., solid line –––) with a corresponding isotype control antibody staining for cells from virus infected (shaded) and peptide immunized (stippled) donors. Data are representative of 3 exp. using 3 mice/exp. C. Total Rb expression levels in CL-4 T cells analyzed in panel 7B with the corresponding isotype control antibody staining.