Replicative history marks transcriptional and functional disparity in the CD8+ T cell memory pool

Abstract

Clonal expansion is a core aspect of T cell immunity. However, little is known with respect to the relationship between replicative history and the formation of distinct CD8⁺ memory T cell subgroups. To address this issue, we developed a genetic-tracing approach, termed the DivisionRecorder, that reports the extent of past proliferation of cell pools in vivo. Using this system to genetically 'record' the replicative history of different CD8⁺ T cell populations throughout a pathogen-specific immune response, we demonstrate that the central memory T (T_CM) cell pool is marked by a higher number of prior divisions than the effector memory T cell pool, owing to the combination of strong proliferative activity during the acute immune response and selective proliferative activity after pathogen clearance. Furthermore, by combining DivisionRecorder analysis with single-cell transcriptomics and functional experiments, we show that replicative history identifies distinct cell pools within the T_CM compartment. Specifically, we demonstrate that lowly divided T_CM cells display enriched expression of stem-cell-associated genes, exist in a relatively quiescent state, and are superior in eliciting a proliferative recall response upon activation. These data provide the first evidence that a stem-cell-like memory T cell pool that reconstitutes the CD8⁺ T cell effector pool upon reinfection is marked by prior quiescence.

Extended Data Fig. 1. Simulation of different scenarios of memory T cell formation.

Simulated data depicting a responding antigen-specific T cell population (blue), comprised of T_EFF undergoing clonal expansion and subsequent contraction (red), plus memory precursor T cells (MP, green) that develop into T_M. Activated T_EFF are modeled to divide rapidly for 6 days (expansion phase), die at a fixed rate throughout the response, and can differentiate into MP cells only during the expansion phase. Cell numbers (top row) and DR^RFP percentages (bottom row) are shown for 3 scenarios: (left) T_EFF can give rise to MP cells during the entire expansion phase, irrespective of the number of prior divisions, (middle) only T_EFF that have gone through at most 24 divisions can give rise to MP cells, or (right) only T_EFF that have gone through at most 10 divisions can give rise to MP cells. Note the strong decay in DR^RFP percentage that is observed during memory formation in case T cell memory is founded by T cells that have undergone few divisions. See Supplementary Note 3 for detailed description and equations.

Extended Data Fig. 2. Evaluation of the division history of T cell subsets throughout a response to Lm-OVA.

a, Gating strategy used to identify indicated T_M populations (d86) in spleen samples. b, DR^RFP percentages within splenic T_M populations (n=6 mice) as identified in panel a. c, DR^RFP percentages within the CD27^HIKLRG1^LO T_CM subset in spleen and lymph nodes (LN) and within the CD27^LOKLRG1^HI T_EM subset in spleen. d, Cell surface expression of CX3CR1, CD62L, and CD43 within splenic CD27^LOKLRG1^HI and CD27^HIKLRG1^LO populations at the peak of the T_EFF phase (day 6 post infection) and in memory phase (day 86 post infection). e, Moving-average of surface marker expression of splenic DR⁺ OT-I T cells during effector phase (day 6), depicted as in Fig. 3g. f, Boxplots depicting DR^RFP percentages within T_EFF (day 6 post infection) subsets in spleen (n=6 mice), relative to the total DR^RFP percentage. g, Kinetics of DR^RFP percentages within CD27^LOKLRG1^HI (left) and CD27^HIKLRG1^LO (right) DR⁺ OT-I T cell populations in blood. Values are relative to the percentage of DR^RFP cells detected at the peak of the response (day 6). Grey lines represent individual mice (n = 22), red and blue lines indicate group mean. h, Simulation of the phenotype model (See Supplementary Note 5 for details) illustrating a scenario in which conversion of CD27^HIKLRG1^LO to CD27^LOKLRG1^HI cells occur only after the peak of the response at a low rate. Depicted are the overall cell numbers (left), and the percentage DR^RFP cells of DR⁺ OT-I T cells (right) in CD27^HIKLRG1^LO cells (blue), CD27^LOKLRG1^HI cells (red) and the total T cell population (green). Note that in this scenario the fraction DR^RFP within the terminally differentiated CD27^LOKLRG1^HI population would increase to almost twice the experimentally observed frequency. All depicted data are representative of at least two independent experiments. Boxplots (c, d, g) represent group median and 25^th/75^th percentiles, whiskers indicate the interquartile range multiplied by 1.5 (c, d) or min/max (g), dots indicate individual samples. P values were determined by one-way ANOVA followed by Tukey’s HSD post-hoc test (c and d), two-sided Student’s T test (c), two-sided repeated measurement correlation test (h), or two-sided Friedman test (g). All significant (< 0.05) P values are indicated in the plots.

Extended Data Fig. 3. Single cell mRNA sequencing of DivisionRecorder⁺ and unmodified memory T cells.

Single cell mRNA sequencing was performed on DivisionRecorder modified and unmodified OT-I memory T cells (Day 75 and 85 post Lm-OVA infection), isolated from spleens (n=7 mice with DR+ memory T cells; n=4 with unmodified memory T cells). Obtained data were aggregated from two independent experiments (Experiment 1: M1-3; Experiment 2: M4-11). All cells were jointly analysed and clustered. a, Cell count per sample. b, Total cell count per MC. c, Sample composition of each MC. d, Relative contribution of DR^GFP and DR^RFP to the total DR⁺ pool within each MC.

Extended Data Fig. 4. T_CM transcriptional states are preserved in DR⁺ OT-I T cells.

Comparison of transcriptional states of splenic memory T cells generated by either DivisionRecorder modified, or unmodified OT-I T cells (Day 75 and 85 post Lm-OVA infection). a-b, Memory OT-I T cells cluster into T_CM (blue) and T_EM (red). 2D projection colored by subset (a), and violin plots depicting normalized UMI counts of selected genes (b) are shown. c, 2D projection of either DR⁺ (left) or unmodified (right) memory OT-I T cells. d, Contribution of DR⁺ and unmodified memory T cells to the T_CM and T_EM subsets. e, Contribution of DR⁺ and unmodified OT-I T cells to the 19 MCs that jointly make up the T_CM subset. Dots indicate individual mice (n=3 per condition). Note that all T_CM states are generated in near-equal proportions by DR⁺ and unmodified memory T cells. Depicted scRNAseq data was obtained from 6 individual mice, and was aggregated from 2 independent experiments. P values were determined by two-sided Student’s T test followed by Bonferroni correction for multiple testing (d and e). P values < 0.05 are indicated.

Extended Data Fig. 5. Replicative history identifies distinct transcriptional states within the T_CM pool.

Single cell transcriptomic profiling of DR⁺ T cells obtained from spleen in memory phase (Day 75 and 85 post Lm-OVA infection). a, Log2 enrichment of selected genes in each MC cluster. Boxplots indicate group median and 25^th/75^th percentiles, whiskers indicate the interquartile range multiplied by 1.5, dots signify individual MCs. The phenotype clusters T_EM, T_CM(eff) and T_CM(mult) contain 4, 9 and 10 MCs, respectively. For definition of T_CM(eff) and T_CM(mult), see Fig. 4B. b, Top and bottom marker genes of ldT_CM (Top, MC2, 11, 14) and hdT_CM (Bottom, MC6, 8, 18), see Fig. 4D for ldT_CM and hdT_CM definitions. c, Heatmaps depicting z-score transformed enrichment values of genes related to cell survival (left), cytotoxicity and effector function (middle), inhibitory markers (top-right), and transcription factors involved in T cell multipotency (bottom-right). Expression is depicted for the 3 ldT_CM and 3 hdT_CM MCs. d, Volcano plot depicting differentially expressed genes in ldT_CM versus hdT_CM. Significantly (adjusted P-value < 0.05) differentially expressed genes are depicted in red. Selected genes are highlighted. e, Cytokine release of CD27^HIKLRG1^LO DR⁺ T cells (isolated from spleen at day >60 post infection) 4 hours post ex vivo stimulation. Percentage DR^RFP cells within cytokine producers (+) and non-producers (-), relative to the average DR^RFP percentage within each sample, is depicted. Lines connect individual ex vivo stimulated samples samples (n=12), obtained from 3 mice. f, Ex vivo degranulation of CD27^HIKLRG1^LO DR⁺T cells (isolated from spleen at day >60 post infection) 4 hours post ex vivo stimulation. Percentage DR^RFP cells within the CD107a/b positive (+) or negative (-) cell populations is depicted. Lines connect individual samples ex vivo stimulated samples (n=17), obtained from 5 mice. g, Enrichment of gene signatures from MSigDB (Hallmark) by gene set enrichment analysis comparing ldT_CM and hdT_CM. Data depicted was accumulated in two independent experiments (3-4 mice per experiment). P values were determined by Tukey’s HSD test (a), Wilcoxon Rank Sum test with Bonferroni correction (d), two-sided Wilcoxon signed-rank test (e, f), the FGSEA algorithm followed by the Benjamini-Hochberg procedure (g). P values < 0.05 are indicated.

Extended Data Fig. 6. gp33-specific P14 T_CM with increased expression of genes associated with replicative quiescence resemble OT-I ldT_CM.

Re-analysis of scRNAseq profiled splenic of P14 memory T cells, published in Kurd et al. (Kurd et al., Science Immunology, 2020). a-b, 2D projection of P14 memory T cells 90 days post LCMV infection, colors indicate individual MCs (a), or the relative expression of effector- and multipotency-associated genes (b). Gene list in Supplementary Table 1. c, P14 memory T cells cluster into T_CM (blue) and T_EM (red). 2D projection colored by subset (top), and violin plots depicting normalized UMI counts of selected genes (bottom) are shown. d, QstemScore of all T_CM MCs in the Kurd et al. dataset. e, Pearson correlations between the Kurd et al. P14 T_CM MCs that score high (MC1, 3) or low (MC6, 7) for QstemScore, and all OT-I T_CM MCs described here. Data are depicted as waterfall plots, asterisks indicate significant correlations. T_CM(eff), T_CM(mult), ldT_CM and hdT_CM MCs are defined in Figure. 4. P values were determined by two-sided Pearson correlation test followed by Bonferroni correction (e). P values < 0.05 are indicated in the plots.

Extended Data Fig. 7. Single cell mRNA sequencing analysis of highly divided and less divided splenic T_CM.

a, Volcano plot depicting differentially expressed genes in Div0-2 versus Div5+ T_CM. Significantly differentially expressed genes (Adjusted P < 0.05) are depicted in red. Selected immune-related genes are highlighted. b, Cell count per MC. c, Number of sequenced cells per sample included in the analysis. d, Sample composition of each MC. e, 2D projection, colors indicate different MCs.Depicted scRNAseq data was collected from 4 individual mice. P values were determined by Wilcoxon Rank Sum test with Bonferroni correction (a).

Extended Data Fig. 8. Modelled T cell responses are consistent with the presence of a replication-competent quiescent T_CM population.

a, Cartoon of the phenotype model depicting phenotypes, the considered interactions among them and the parameters associated with the interactions. Arrows indicate various events occurring during the response, such as cell division (denoted with λ), differentiation to a different phenotype (denoted with δ), cell death during contraction (denoted with μ), and recruitment toward the secondary response during recall infection (denoted with r). Subscripts indicate the phenotype of the cell that the parameter is affecting. Full list of parameters can be found in Supplementary Note 5. b-d, Best fit of the modelled T cell response to the experimental measurements depicting either cell numbers (top plot in each panel), or DR^RFP percentages (bottom plot in each panel). The total number of quiescent T cells generated was either capped at 1% (b) or 0.1% (c, d) of the T_EFF pool. Lines depict the modeled populations; Dots indicate the experimental measurements obtained from peripheral blood (b, d) or spleen (c). See Supplementary Note 5 for more details and calculations. Experimental data points are representative of at least two independent experiments, dots indicate individual mice (n=6 mice per time point).

Extended Data Fig. 9. Model describing replicative behaviors in the CD8 ⁺memory T cell pool.

Upon infection, antigen-specific CD8⁺ T cells activate and rapidly expand (phase 1, p1). Following pathogen clearance (p2), a subset of memory T cells continues to divide, resulting in a progressive increase in the replicative history of the overall T cell memory pool (dotted line). Within this population, three separate behaviors of transcriptionally disparate memory T cell pools can be distinguished. Top) Terminally differentiated T_EM cells that cease division after the inflammation phase (p1) and that are marked by high transcription of effector- and minimal expression of multipotency-associated genes ([E], [M]). Upon reactivation, these cells exert rapid effector functions, but lack the potential to re-expand. Middle) A subgroup of T_CM that continues to proliferate in the memory phase, exhibits diminished levels of multipotency-associated transcripts, and that abundantly expresses effector-associated genes. Although the functionality of these cells upon reinfection requires further study, their heightened expression of effector-associated genes suggests that these cells exert cytotoxic activity upon reinfection. The contribution of these cells to the secondary T_EFF pool is limited. Bottom) A subgroup of T_CM cells that shows low expression of effector-associated genes but increased expression of multipotency-associated genes, and that exists in a near-quiescent state after the inflammation phase. Upon renewed infection, this cell pool is primarily responsible for the generation of a new wave of secondary T_EFF. Based on their transcriptional profile, these cells are expected to have limited immediate cytotoxic functions.

Extended Data Fig. 10. Gating strategy.

General gating applied to flow cytometry data presented in the study. Single lymphocytes were first selected using morphology gates, and were subsequently gated on CD8⁺ T cells and transferred OT-I T cells (Vβ5⁺CD45.2⁺). Next, DR^RFP and RF^GFP could be directly selected, or first separated by phenotype depending on the analysis. The data presented here was analyzed from blood of a recipient of DR⁺ cells, and was acquired 6 days post infection with Lm-OVA. Phenotype gates other than those shown here are defined in their respective figures.

Fig. 1. DivisionRecorder activation is a proxy for replicative history.

a, Schematic overview of the DivisionRecorder system. b, Cartoon depicting progressive DivisionRecorder activation in a proliferating cell pool. c, Simulation of the minimal ODE model (See Supplementary Note 2 for detailed description and equations), depicting DR^RFP acquisition as a function of population doublings for the indicated values of DR^RFP acquisition probability (p). d, Maximal number of theoretically recordable population doublings, approximated by calculating the amount of division events required to reach a 99% DR^RFP population. Approximate maximums for selected values of p are indicated, colors correspond to legend in panel c. e-f, Percentage of DR^RFP cells over time in cultured DivisionRecorder⁺ (DR⁺) CRE-activity reporter HEK 293T cells (n=3 replicates per group) in which the CRE recombinase gene was preceded by either a stable nucleotide region (indicated as “no STR”) or a repeat of 24 guanines (indicated as “with STR”). Representative plots (e) and summarizing line graphs (f) are shown. Lines connect experimental replicates g, Percentage of DR^RFP cells across population doublings in DR⁺ CRE-activity reporter HEK 293T cells (n=3 replicates per group) in which the CRE recombinase gene was preceded by either a low stability STR ([G]24) or a high stability STR ([CA]30). Dots indicate individual samples, lines represent fitted linear regression, dotted lines indicate bounds of the 95% confidence interval. h-i, Percentage of DR^RFP cells across population doublings in immortalized DR⁺ mouse embryonic fibroblasts. Representative flow cytometry plots (h) and summarizing graph (i) are shown. Best fits of the minimal ODE model are depicted (100 bootstraps per experimental replicate, Supplementary Note 2). Blue line represents the median of the bootstraps, grey lines represent individual fits, dots indicate experimental measurements (n=3 replicates). p indicates the estimated DR^RFP acquisition probability. Depicted experimental data are representative of at least two independent experiments. P values (g) were determined by two-sided ANCOVA test.

Fig. 2. The DivisionRecorder can be applied to study T cell division kinetics in vivo.

a, Overview of experimental setup. b-c, DR⁺ OT-I T cells were transferred into recipient mice 24 hours post infection with Lm-OVA. Spleen samples were analyzed for the percentage of DR^RFP cells at day 1-4 post cell transfer. Representative pseudo-color density plots (b), and boxplots (c) in which the boxes indicate group median and 25^th/75^th percentiles, whiskers represent min/max, dots represent individual samples (n=8 mice for day 1 and 2; n=7 mice for day 3 and 4). d-e, CTV-stained OT-I T cells were retrovirally transduced with the DivisionRecorder and transferred into recipient mice (n=4) 24 hours post infection with Lm-OVA. 48 hours post-transfer, splenic DR⁺ OT-I T cells were assessed for CTV dilution (d), and the percentage of DR^RFP cells within each division peak was analyzed (e). All depicted data are representative of at least two independent experiments, lines and symbols indicate individual mice or samples. P values were determined by two-sided Kruskal-Wallis test, with Dunn’s multiple comparisons test (c), or two-sided repeated measurement correlation test (e).

Fig. 3. The multipotent memory T cell pool is formed by replicative ‘mature’ cells.

a-c, Kinetics of DR⁺ OT-I T cells (a) and the percentage of DR^RFP relative to day 4 (c) in response to Lm-OVA, measured in peripheral blood (n=6 mice). Representative flow cytometry plots (b) showing DR^RFP and DR^GFP frequencies at indicated time points, and line graphs (a, c) depicting kinetics of single mice (grey) and group median (black). d, DR^RFP percentages within blood at day 5/6 (T_EFF) and day >60 (TM) following LCMV-OVA infection (n=7). e, Representative plots depicting DR^RFP frequencies in blood (Bl), spleen (Spl) and liver (Liv). f, Percentage of DR^RFP detected in indicated organs of recipient mice at the indicated time points (n=6 mice per time point; response to Lm-OVA). Boxplots indicate group median and 25^th/75^th percentiles, whiskers represent min/max, dots represent individual samples. g, Moving average of surface marker expression level on splenic DR⁺ cells plotted against the percentage of DR^RFP within each window during memory (day 86; n=6), means are shown in black. DR^RFP percentages within each window are corrected for the total percentage of DR^RFP detected in that sample. h, Gating strategy (left) and DR^RFP percentages (right) of CD27^HIKLRG1^LO and CD27^LOKLRG1^HI cells in spleen during effector (d6, top) and memory phase (d86, bottom; n=6) in response to Lm-OVA. i, DR^RFP percentages within the CD27^HIKLRG1^LO and CD27^LOKLRG1^HI cell populations in blood, comparing effector (day 5/6) and memory (day >60) phases. Data shown for Lm-OVA (top; n=22) and LCMV-OVA (bottom; n=7) infections. Lines connect individual mice. j-k, Ki67 expression by CD27^HIKLRG1^LO and CD27^LOKLRG1^HI OT-I cells in blood in response to Lm-OVA. Representative flow cytometry plot (j), and line graphs (k) where solid lines indicate population means, shaded areas indicate 95% confidence interval (n=11 mice). All depicted data are representative of at least two independent experiments, lines and symbols indicate individual mice or samples. P values were determined by two-sided Kruskal-Wallis test with Dunn’s multiple comparisons test (f), or two-sided Wilcoxon’s signed-rank test (d, h, i).

Fig. 4. Replicative history identifies distinct transcriptional states within the T_CM pool.

Single cell transcriptomic profiling of DR⁺ T cells obtained from spleen in memory phase (Day 75 and 85 post Lm-OVA infection). a, 2D projection of all profiled cells, colors indicate MCs (left), or relative expression of effector- and multipotency-associated genes (right). Gene list in Supplementary Table 1. b, Hierarchical clustering of MCs by their expression of effector- and multipotency-associated genes used in A. MCs are divided into 3 clusters based on Euclidean distance. c, Expression of selected genes by each MC cluster. d, DR^RFP/DR^GFP ratio within each MC, depicted as waterfall plot (left) and boxplot (right). e-f, Enrichment of gene signatures from MSigDB (C7, collections deposited by Goldrath (GR) and Kaech (KA), Supplementary Table 2) by gene set enrichment analysis comparing ldT_CM and hdT_CM (e), and enrichment plots (f) of 2 representative gene sets. g, Heatmaps depicting genes involved in immune function that were significantly (P < 0.05) depleted (left) or enriched (right) within ldT_CM (See Extended Data Figure 4d, Supplementary Table 3). Selected genes are annotated, complete gene lists in Supplementary Table 4. h, QstemScore of all T_CM MCs depicted as waterfall plot (left) and boxplot (right). QstemScore is based on marker genes of quiescent stem cells (Supplementary Table 5), see methods for calculation. Data depicted were accumulated in two independent experiments (3-4 mice per experiment). Boxplots (c, d, h) indicate group median and 25^th/75^th percentiles, whiskers indicate the interquartile range multiplied by 1.5, dots signify individual MCs. The phenotype clusters T_EM, T_CM(eff) and T_CM(mult) contain 4, 9 and 10 MCs, respectively. P values were determined by two-sided Tukey’s HSD test (c), two-sided Student’s T test with false-discovery rate correction (d, h), the FGSEA algorithm followed by the Benjamini-Hochberg procedure (e), or two-sided Wilcoxon Rank Sum test with Bonferroni correction (g). Significant P values (< 0.05) are indicated in the plots.

Fig. 5. Replicative history is linked to recall-potential within the T_CM pool.

a, Experimental setup. Primary recipient mice received 5*10⁵ naïve OT-I and 5*10⁵ naïve GFP;OT-I T cells. 30 days after Lm-OVA challenge, CD8 T cells were enriched, labelled with CTV and transferred into infection-matched secondary recipient mice (1 primary recipient per secondary recipient). At d105 post infection, splenic CD27⁺KLRG1^- memory T cells that had either divided 0-2 or 5+ times and were either GFP⁺ or GFP- were isolated by FACS. b, Enrichment of gene signatures from MsigDB (C7, collections deposited by Goldrath (GR) and Kaech (KA), Supplementary Table 2) between Div0-2 and Div5+ cells. Top and bottom 5 pathways are depicted. c, Enrichment plots of representative pathways detected in by gene set enrichment analysis. d, Ratio of normalized counts between Div0-2 and Div5+ cells within each MC separately calculated for GFP⁺ and GFP^- populations. Bars indicate averages, dots indicate ratios of either GFP⁺ or GFP^- OT-I T cells. Red dotted lines indicate a fold change of 2. e, Waterfall plots depicting top and bottom 6 marker genes for selected MCs, filtered for genes involved in immune function (Supplementary Table 3). f, Flow cytometry plots depicting pre-transfer mixes of Div0-2 and Div5+ T_CM. g, 8,000-12,000 total memory T cells as described in f were transferred into infection-naïve mice, following Lm-OVA challenge 24 hours later. Ratios between Div0-2 and Div5+ derived cells was determined from peripheral blood samples at indicated days post infection. Lines connect populations from individual mice (Experiment 1 n = 3; Experiment 2 n = 5). Depicted scRNAseq data was collected from 4 mice, data describing recall potential was obtained from 8 mice. P values were determined by the FGSEA algorithm followed by the Benjamini-Hochberg procedure (e).

Fig. 6. The secondary T_EFF pool is predominantly generated by previously quiescent memory T cells.

a, Kinetics of the percentage of DR^RFP cells in blood upon secondary Lm-OVA infection. Values are relative to the DR^RFP percentage within the respective memory pools (n=6 mice), black line represents group mean. b-c, DR^RFP percentages in indicated organs (b), or within splenic CD27^LOKLRG1^HI and CD27^HIKLRG1^LO populations (c) at indicated time points (n=6 mice per time point) post-secondary infection. Boxplots indicate group median and 25^th/75^th percentiles, whiskers represent min/max, dots represent individual samples. d, DR^RFP percentages in blood at memory (day >60) and at the peak of the secondary response (day 4/5 post-recall). Memory pools were generated with LCMV-OVA, recall infection was performed with Lm-OVA. e, DR^RFP acquisition in blood following primary and secondary infection. Values are relative to DR^RFP percentage at the peak of the primary or secondary response. Lines represent group medians (n=6 mice per group), greyed areas represent 95% confidence intervals. f, DR^RFP percentages in blood during effector and memory phases of the primary and secondary responses. Lines connect data of individual mice (n=6). g, DR^RFP percentages in blood (n=5 mice) upon tertiary infection. Mice were challenged twice with Lm-OVA with a >60 days interval, and subsequently infected with LCMV-OVA >60 days post-secondary infection. Depicted data are representative of at least 2 independent experiments. P values were determined by two-sided Kruskal-Wallis test with Dunn’s multiple comparisons test (b, c), two-sided Wilcoxon signed-rank test (d, f), or repeated-measures one-way ANOVA followed by Dunnett correction (g).

Fig. 7. Modelled T cell responses are consistent with the presence of a replication-competent quiescent T_CM population.

a, Division history of T_CM and T_EM pools generated by modelled T cell responses (see Supplementary Note 5) during which a high (capped at 1% of the T_EFF pool size) or low (capped at 0.1% of the T_EFF pool size) fraction of T cells acquire quiescence during the effector phase (top). 3 re-expansion functions were used to restrict which fraction of T_CM with a given number of prior divisions will re-expand during recall (bottom). For reference, the division history of T_CM is shown as a shaded area. b, Modelled DR^RFP percentages within the CD27^LOKLRG1^HI and CD27^HIKLRG1^LO populations during secondary responses, with each re-expansion function applied to a memory pool containing either a high or low number of quiescent T_CM. Black dots indicate experimental measurements. j, Best fit of the modelled T cell response (number of quiescent T cells capped to 1% of T_EFF) experimental data obtained from spleen, depicting either cell numbers (left) or DR^RFP percentages (right). See Supplementary Note 5 for details. Lines indicate the modeled populations; dots indicate experimental measurements.