Enhancing thymic function improves T-cell reconstitution and immune responses in aged mice

Abstract

Age-related thymic involution leads to diminished output of naïve T-cells. While this process is suggested to increase the risk of disease severity in the elderly following infection, direct evidence is lacking. We developed two mouse models that allow us to experimentally prevent or reverse thymic involution. Constitutive Myc expression in thymic epithelial cells (TEC) of middle-aged mice enhanced thymic function, and increased numbers of peripheral naïve CD4 and CD8 T-cells. Inducible Myc expression reversed age-related thymic involution and partially recovered peripheral naïve T-cell numbers. Importantly, improving thymic function in these settings preserved T-cell-dependent antibody responses and significantly reduced T-cell-associated mortality after infection with Toxoplasma gondii. Improved thymic function also rebalanced age-associated alterations in the Treg pool, and mitigated loss of the transcriptional Th1 signature in aged conventional T-cells. Our findings support the value of TEC-focused thymic regeneration strategies for enhancement of T-cell-mediated immunity in the elderly.

Fig 1. Enhanced thymic function overcomes T-cell-mediated mortality in middle-aged individuals infected with T. gondii.

Mice were infected with 10 cysts of T. gondii (strain ME49) and survival was followed up for 30 days. (A) Kaplan–Meier plots showing the survival of WT young (2–3 months of age) vs. WT middle-aged (8–12 months of age) vs. WT aged mice (15-18 months of age) (n = 11–24 mice per group from three independent experiments). (B) The schematic on the top shows the αCD4 and αCD8 administration protocol. Kaplan–Meier plots showing the survival of WT young vs. WT middle-aged mice (left) (n = 5-8 mice per group from two independent experiments) and WT young vs. WT aged mice (right) (n = 7–8 mice per group from two independent experiments), receiving depleting antibodies (dotted lines) or PBS (solid lines). (C) A simplified schematic of the genetic mouse model and the experimental groups. β5t-Cre mice [30] were crossed to Myc^stopFL mice [31] to generate the constitutive Myc transgenic model (MycTg, left). A timeline showing the age intervals of mice selected for experimental groups (right). (D) Representative pictures (left) and total thymus cell counts (right) of WT young, WT middle-aged, and MycTg middle-aged mice. (E) H&E staining of thymus sections from 9-week-old WT young (left), 6-month-old WT (middle), and 6-month-old MycTg (right) mice. (C: cortex, M: medulla). Scale bar = 200 µm. (F) Kaplan–Meier plots showing the survival of WT young vs. WT middle-aged vs. MycTg middle-aged mice (n = 6–12 mice per group from two independent experiments). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 2. Enforced expression of Myc in TEC.

(A) Representative FACS plots showing total thymic epithelial cells (TEC, CD45⁻ EpCAM⁺). Bar graphs displaying TEC percentages and absolute cell numbers in WT young (2–3 months of age), WT middle-aged (8–12 months of age), and MycTg middle-aged (8–12 months of age) mice. (B) Representative FACS plots showing Ly51 and UEA1 staining on TEC. Bar graphs indicate percentages and absolute cell numbers of the Ly51⁻ UEA1⁻ TEC, Ly51⁺ UEA1⁻ (cTEC) and Ly51⁻ UEA1⁺ (mTEC) subsets within the total TEC, and the cTEC:mTEC ratio. (C) Representative FACS plots showing MHCII and CD80 expression within mTEC. Bar graphs indicate percentages and absolute cell numbers of MHCII^hi CD80^hi (mTEC^hi), MHCII^lo CD80^lo (mTEC^lo) subsets within mTEC. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 3. Enforced Myc expression in TEC prevents the age-associated decline in peripheral naïve CD4 and CD8 T-cells.

(A) Total spleen cellularity of WT young (2–3 months of age), WT middle-aged (8–12 months of age), and MycTg middle-aged (8–12 months of age) mice. (B) Representative FACS plots of TCRβ expression on total splenic cells. Bar plots showing the percentages and absolute numbers of TCRβ⁺ cells. (C) Representative FACS plots and bar plots of the CD4 and CD8 T-cells (pre-gated in TCRβ⁺ cells) frequencies and absolute numbers. (D) CD62L and CD44 expression on TCRβ⁺ CD4 (top) and TCRβ⁺ CD8 (bottom) T-cells. Bar plots depicting percentages and quantitation of CD62L⁺ CD44⁻ (naive), CD62L⁺ CD44⁺ central memory (CM), and CD62L⁻ CD44⁺ effector memory (EM) in WT young, WT middle-aged, and MycTg middle-aged mice. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 4. Myc-mediated TEC expansion prevents age-associated alterations to peripheral T-cells, including Treg accumulation, reduced TCR repertoire diversity, and T-cell-mediated antibody responses.

(A) Representative Foxp3 staining in TCRβ⁺CD4 T-cells in the spleens of WT young (2–3 months of age), WT middle-aged (8–12 months of age), and MycTg middle-aged (8–12 months of age) mice. (B) Representative FACS plots of the CD25 expression within Foxp3⁺ Treg. Bar plots showing the CD25⁺ Treg frequency. FACS histograms and bar plots showing the median fluorescence intensity (MFI) of CD25 pre-gated in Foxp3⁺ Treg. Dot plots showing the absolute number of CD25⁺ vs. CD25⁻ Treg per mouse in WT middle-aged and MycTg middle-aged relative to WT young mice. (C) The schematic of the αCD25 administration protocol (left). Kaplan–Meier plot showing the survival curve of WT young vs. WT middle-aged mice that received αCD25 depleting antibodies or vehicle (right) (n = 9–13 mice from two independent experiments). (D) Simpson’s clonality index of CD8 T-cells isolated from the spleens of WT young and WT aged mice (left) or WT young, WT middle-aged and MycTg middle-aged (right). (E) NP-specific IgG quantification in serum at day 14 after immunization of the indicated mouse groups. *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 5. Thymus regeneration delays the mortality of aged mice infected with T. gondii.

(A) A simplified schematic illustrating the experimental model and the experimental groups used. β5t-rtTA × tetO-Cre mice [44] were crossed to Myc^stopFL [31] to generate iMycTg mice (left). Mice at 9–12 months of age were given doxycycline in drinking water for 2 weeks and used in experiments 3–6 months later (averages and ranges of the elapsed time post-doxycycline are given below for each experiment). WT aged (15–18 months of age) mice were given doxycycline as iMycTg mice. WT mice at 2–3 months of age were also included as young controls. (B) Representative pictures (left) and total thymus cellularity (right) of the WT young, WT aged, iMycTg-U (unenlarged), and iMycTg-E (enlarged) mice. Mice were analyzed on average 4.3 months after induction (range: 3.5–5.1 months). (C) Schematic of doxycycline induction and T. gondii infection (left). Kaplan–Meier plot showing the survival of WT young vs. WT aged vs. iMycTg aged-U vs. iMycTg aged-E mice (right) after infection. The experiment was concluded on day 21 post-infection as iMycTg-E aged mice were moribund. Mice were analyzed on average 3.1 months after induction (range: 3–3.6 months) (n = 5–10 mice per group). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 6. Inducible enforced expression of Myc in TEC.

(A) Representative FACS plots showing total thymic epithelial cells (TEC, CD45⁻ EpCAM⁺) in live cells. Bar graphs indicate TEC percentages and absolute cell numbers in WT young (2–3 months of age), WT aged (15–18 months of age), iMycTg aged-U (unenlarged, 15–18 months of age), and iMycTg aged-E (enlarged, 15–18 months of age) mice. (B) Representative FACS plots showing Ly51 and UEA1 detection pre-gated in TEC. Bar graphs indicate the absolute cell numbers and percentages of Ly51⁻ UEA1⁻, Ly51⁺ UEA1⁻ (cTEC), and Ly51⁻ UEA1⁺ (mTEC) subsets within total TEC. (C) Representative FACS plots showing MHCII and CD80 expression within mTEC. Bar graphs indicate percentages and absolute cell numbers of MHCII^hi CD80^hi (mTEC^hi) and MHCII^lo CD80^lo (mTEC^lo) subsets within mTEC. Mice were analyzed on average 3.9 months after induction (range: 3.5–4.8 months). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 7. Thymus regeneration promotes naïve T-cell recovery in iMycTg aging-E mice and a partial decrease in Treg frequencies.

(A) Total spleen cellularity of WT young (2–3 months of age), WT aged (15–18 months of age), and iMycTg aged-E (enlarged) (15–18 months of age) mice. (B) Representative FACS plots of the expression of TCRβ in the total splenic cells. Bar plots showing the percentages and absolute numbers of TCRβ⁺ cells. (C) Representative FACS plots and bar plots of the CD4 and CD8 T-cell (pre-gated in TCRβ⁺) frequencies and absolute numbers in the indicated mouse groups. (D) CD62L and CD44 expression on TCRβ⁺ CD4 (top) and TCRβ⁺ CD8 (bottom) T-cells. Bar plots depicting percentages and quantitation of CD62L⁺ CD44⁻ (naive), CD62L⁺ CD44⁺ central memory (CM), and CD62L⁻ CD44⁺ effector memory (EM) T-cells in the indicated mouse groups. (E) Representative Foxp3 staining on CD4 splenic T-cells. Bar plots show the percentage and quantification of Foxp3⁺ Treg cells. (F) FACS plots and bar plots of the CD25⁺ cells frequency within Foxp3⁺ Treg in WT young, WT aged and iMycTg aged-E. Histograms and bar plots in the middle show the median fluorescence intensity (MFI) of CD25 pre-gated in FoxP3⁺ Treg, relative to WT young mice. Dot plots on the right show the absolute number of CD25⁺ and CD25⁻ Treg cells per mouse in the indicated mouse groups. Mice were analyzed on average 4.5 months after induction (range: 3.4–6.0 months). *p < 0.05, **p < 0.01, ***p < 0.001, and ****p < 0.0001. The data underlying this figure can be found in S7 File.

Fig 8. Thymus regeneration increases TCR repertoire diversity and partially restores transcriptional signatures of T-cells during aging.

(A) Simpson’s clonality index of CD8 T-cells isolated from the spleens of WT young (2–3 months of age), WT aged (15–18 months of age), and iMycTg aged-E (enlarged) (15–18 months of age) mice. Mice were analyzed 5.9 months after induction. (B) Overview illustrating single-cell (sc)-RNA Seq experiment on T-cells. Data from this study were integrated with two other publicly available datasets [47,48] to ensure robustness, but only cells from this study are displayed in the figure. (C) Annotated UMAP of T-cells isolated from WT young, WT aged, and iMycTg aged-E mice at steady state. Annotated populations are shown at the top, and cell densities per experimental group are shown at the bottom. (D) Heatmap of the annotated populations. (E) Bar plots showing the abundance of CD4 and CD8 T-cells subsets identified by scRNA-Seq. (F) Heatmaps on genes upregulated in aged CD4 (top) and CD8 (bottom) T-cells used to make an aging score (see Materials and methods) shown in G. (G) Violin plots showing aging scores for naïve, effector memory (EM) and regulatory T-cells (Treg) CD4 T-cells (top), and for naïve, CM and EM CD8 T-cells (bottom). (H) UMAPs (left) and bar plot (right) showing Treg (Foxp3⁺) cells grouped by their expression of Il2ra (encoding CD25) and Pdcd1 (encoding PD1). Mice were analyzed 5.9 months after induction. The data underlying this figure can be found in S7 File.

Fig 9. Transcriptome analysis reveals an altered Th1 signature in aged mice infected with T. gondii.

(A) Overview of the scRNA-Seq experiment; WT young (2–3 months of age), WT middle-aged (8–12 months of age), MycTg middle-aged, WT aged (15–18 months of age), and iMycTg aged-E (enlarged) (15–18 months of age) mice were infected with 10 cysts of T. gondii. T-cells were isolated from the spleen on day 12 after infection. A UMAP (B) and a heatmap (C) of the annotated CD4 and CD8 T-cell populations. (D) Bar plot showing the abundance of the annotated CD4 and CD8 T-cell populations. (E) Heatmap showing the expression of 42 of the total Th1-associated genes [63] within the activated CD4 T-cell cluster. Mice were analyzed 5.9 months after induction. The data underlying this figure can be found in S7 File.