Fig 1.
Neonatal exposure to CMV drives a robust memory inflation response.
(A) Schematic of experimental design. Newborn mice were infected with MCMV-gB at birth. CD8 + T cells were isolated from the spleen at 1, 2, 4, 8, 12, and 21 weeks post birth. (B) CD8 + T cells were stained with gB tetramer-specific cells and examined by flow cytometry (N = 4–10 mice). At 21 weeks, the total number of (C) CD8 + T cells, (D) naïve cells (CD49d- CD44-) and (E) antigen-experienced CD8 + T cells (CD44 + CD49d+) were quantified. Within the tetramer subgates, memory CD8 + T cells were identified as Central Memory (TCM, CD27 + CX3CR1-), Peripheral Memory (TPM, CD27 + CX3CR1+) or Effector Memory (TEM, CD27- CX3CR1+). (F) Representative dot plots of the gating scheme and (G) quantification of each memory subset are represented. For two groups, an unpaired t-test with Mann-Whitney test for correction was performed. For more than two groups, an ordinary one-way ANOVA with Tukey’s multiple comparisons test was performed. Results are shown as mean ± SD or mean only. ***p < 0.001, ****p < 0.0001.
Fig 2.
Neonatal CMV infection recruits and maintains CD8 + T cells closest to thymic egress into the memory inflation response.
(A) Experimental schematic. Newborn mice were infected with MCMV-gB at birth. Uninfected mice were injected with PBS as control. Mice were given tamoxifen at 1 day, 7 days, or 28 days post birth to ‘timestamp’ CD8 + T cells with a Zsgreen fluorescent tag. (B) Mice were bled at 16–17 weeks post birth, and circulating CD8 + T cells were examined for the percentage marked (ZsGreen+) by flow cytometry (N = 6–24 mice). At 16–17 wks post birth, CD8 + T cells from the spleen were enriched and gB peptide stimulation with BFA was performed for 4 hours (N = 19–30 mice). (C) Representative contour plots of CD8 + IFNγ+ cells and (D) statistical analysis of IFNγ production. Spleens collected at 24 weeks post birth were stained for antigen-specific tetramer+ cells (N = 15–20 mice). (E) Representative contour plots and (F) paired statistical analysis of tetramer+ CD8 + T cells within the timestamp+ and timestamp- populations in the spleen. For two groups, an unpaired t-test with Mann-Whitney test for correction was performed. For more than two unpaired groups, an ordinary one-way ANOVA with Tukey’s multiple comparisons test was performed. For paired group analysis, a two-way RM ANOVA with Bonferroni correction was performed. Results are shown as mean ± SD or mean only. **p < 0.01, ****p < 0.0001.
Fig 3.
Age of the CD8 + T cells recruited into the memory inflation response dictates phenotype.
Newborn mice were infected with MCMV-gB at birth. Uninfected mice were injected with PBS as control. Mice were given tamoxifen at 1d, 7d, or 28d post birth to ‘timestamp’ CD8 + T cells with a Zsgreen fluorescent tag. Spleens were collected at 24 wks post birth, and CD8 + T cells were phenotyped by flow cytometry. (A) Representative density plot of CD27 vs. CX3CR1 within the tetramer+ subgates, where memory CD8 + T cells were identified as Central Memory (TCM, CD27 + , CX3CR1-), Peripheral Memory (TPM, CD27 + , CX3CR1+), or Effector Memory (TEM, CD27-, CX3CR1+). (B) Quantification of TCM, TPM, and TEM phenotypes of 1d, 7d, and 28d timestamped mice within the tetramer+ population (N = 15–20 mice). (C) Quantification of the percentage of timestamped CD8 + T cells within different tissues, as indicated. One-way ANOVA with Tukey’s multiple comparisons test was performed. Results are shown as mean. *p < 0.05, **p < 0.01.
Fig 4.
Adult CMV infection recruits and maintains CD8 + T cells closest to thymic egress into the memory inflation response.
(A) Adult mice were infected with MCMV-gB at 56d post birth, and spleens were collected at 4, 8, 12, and 20 wks post infection. (B) Quantification of tetramer+ CD8 + T cells. (C) Experimental schematic. Briefly, adult mice were infected with MCMV-gB at 56 days post-birth. Uninfected mice were injected with PBS as control. Mice were given tamoxifen at 1d, 7d, 28d, and 56d post birth to ‘timestamp’ CD8 + T cells with a Zsgreen fluorescent tag. Mice were bled at 16 wks post birth, and circulating ZsGreen+ CD8 + T cells were examined by flow cytometry. (D) The percentage of CD8 + T cells that are ZsGreen+ positive (N = 5–12 mice). (E) Spleens were collected from adults at >24 wks post birth, and CD8 + T cells were stained for gB tetramer. Representative contour plots and percentage of tetramer+ CD8 + T cells in the timestamped population. For two groups, an unpaired t-test with Mann-Whitney test for correction was performed. Results are shown as mean. *p < 0.05, ***p < 0.001, ****p < 0.0001.
Fig 5.
Recent thymic emigrants are preferentially recruited into the memory inflation response regardless of developmental origin.
(A) Experimental schematic. Newborn thymuses were collected from TdTomato+ timestamp reporter mice and surgically transplanted under the kidney capsule into adult (>8 wk) ZsGreen+ timestamp reporter mice. Mice were administered tamoxifen for 3 d and then infected with MCMV-gB two days after marking. At 4, 8, 12, and 16 wks post infection, mice were bled and the CD8 + T cell memory response was measured by flow cytometry. (B) The percentage of CD8 + cells that were ZsGreen/TdTomato marked cells (N = 13–21 mice). A two-way repeated measures ANOVA with Bonferroni test correction was performed. Results are shown as pair-wise comparison, where each connected point is one mouse. (C) Experimental schematic. Mice were administered tamoxifen at 1d to mark the ‘fetal’ layer or 28d post birth to mark the ‘adult’ layer. All mice were infected at 56d post birth and bled 4, 8, and 12 wks post infection. (D) The percentage of CD8 + T cells from 1d or 28d marked layers. A two-way ANOVA with Bonferroni test correction was performed. *p < 0.05, ***p < 0.001, ****p < 0.0001.
Fig 6.
RTE-derived memory cells adopt a terminally differentiated phenotype.
(A) Experimental schematic. Adult ZsGreen+ mice were marked with tamoxifen 6 wks prior to harvest, while adult Ai9 TdTomato+ mice were marked with tamoxifen 2 wks prior to harvest. CD8 + T cells were isolated from the spleen. A 1:1 ratio of ZsGreen+ and TdTomato+ CD8 + T cells were pooled and co-transferred into adult TCRα-/- animals. The next day, recipient animals were infected with MCMV-gB. Mice were bled at 9 wks post infection, and the CD8 + T cell response was measured by flow cytometry. (B) The percentage of ZsGreen or TdTomato-marked CD8 + T cells that were Tetramer+ (N = 9). (C) Representative 2-way FACS plot of CD27 vs. CX3CR1. (D) The percentage of tetramer+ timestamped CD8 + T cells that adopted a central memory (TCM), peripheral memory (TPM), or effector memory (TEM) phenotype. A paired t-test with Wilcoxon matched pairs signed rank test performed for correction. Results are shown as pair-wise comparison, where each connected point is one mouse. *p < 0.05, ***p < 0.001, ****p < 0.0001.