The initial age-associated decline in early T-cell progenitors reflects fewer pre-thymic progenitors and altered signals in the bone marrow and thymus microenvironments

Abstract

Age-related thymus involution results in decreased T-cell production, contributing to increased susceptibility to pathogens and reduced vaccine responsiveness. Elucidating mechanisms underlying thymus involution will inform strategies to restore thymopoiesis with age. The thymus is colonized by circulating bone marrow (BM)-derived thymus seeding progenitors (TSPs) that differentiate into early T-cell progenitors (ETPs). We find that ETP cellularity declines as early as 3 months (3MO) of age in mice. This initial ETP reduction could reflect changes in thymic stromal niches and/or pre-thymic progenitors. Using a multicongenic progenitor transfer approach, we demonstrate that the number of functional TSP/ETP niches does not diminish with age. Instead, the number of pre-thymic lymphoid progenitors in the BM and blood is substantially reduced by 3MO, although their intrinsic ability to seed and differentiate in the thymus is maintained. Additionally, Notch signaling in BM lymphoid progenitors and in ETPs diminishes by 3MO, suggesting reduced niche quality in the BM and thymus contribute to the early decline in ETPs. Together, these findings indicate that diminished BM lymphopoiesis and thymic stromal support contribute to an initial reduction in ETPs in young adulthood, setting the stage for progressive age-associated thymus involution.

Keywords: aging; early T-cell progenitors; hematopoiesis; thymus involution.

FIGURE 1

Early T‐cell progenitor (ETP) and total thymocyte cellularity declines by 2MO of age. (a) Quantification of total thymocytes and ETPs in C57BL/6J mice at the indicated ages. (b) Linear regression analysis of total thymic cellularity and ETP cellularity at each age. R ² = coefficient of correlation (c) Ratio of DN2 to ETP cellularity at each age. Symbols represent (a, c) data from individual mice and (b) an average of 7–14 mice at each age. Bars represent means ± SEM from 3–4 independent experiments for each age group. Statistical analysis in (a) was performed using one‐way ANOVA with Dunnett's multiple comparisons test where *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 2

Number of available, functional TSP niches does not decline with age. (a) Schematic of the multicongenic barcoding experiment to quantify available niches in mice of different ages. Lin⁻Flk2⁺CD27⁺ BM progenitors were isolated from eight different congenic strains (Figure S2B), mixed at equal ratios and i.v. injected into non‐irradiated 1MO, 3MO, 6MO, and 12MO mice. After 21 days, the thymi from recipient mice were analyzed by flow cytometry to quantify the number of detectable donor strains that underwent T‐cell differentiation. (b) Pie charts illustrate the number and relative frequency of distinct color‐coded donor strains in individual recipients of the indicated ages. (c) The estimated number of available functional niches in each age recipient as determined by multinomial sampling. Symbols indicate data from individual mice; bars represent means ± SEM compiled from four independent experiments (n = 10–12 mice per recipient age).

FIGURE 3

3MO T‐cell progenitors show comparable seeding and differentiation capacity relative to 1MO progenitors. (a–c) Donor chimerism in 1MO and 3MO recipient thymuses 21d after transplantation of an equal number of 1MO and 3MO Lin⁻Flk2⁺CD27⁺ BM progenitors into 1MO or 3MO non‐irradiated recipients. Quantification of the (a) frequency and (b) number of 1MO and 3MO donor‐derived thymocytes in recipients of the indicated ages. (c) Quantification of the percentage of DP thymocytes derived from 1MO and 3MO donors detected in recipient mice of the indicated ages. Data are pooled from four independent experiments (n = 6–10 mice per recipient age). (d) Representative flow cytometry plots and (e) quantification of the percentage of ETPs that incorporated bromodeoxyuridine (BrdU) 8 h after i.p. injection in mice of the indicated ages. Data are compiled from four independent experiments (n = 5 mice per age group). (f) Representative flow cytometry plots and (g) quantification of the frequency of cleaved caspase‐3⁺ cells in the indicated DN thymocyte subsets from 1MO and 3MO mice. Data are pooled from four independent experiments (n = 4 mice per age group). (a–c, e, g) Symbols represent data from individual mice. Bars represent means ± SEM. p value for (a) was obtained by two‐way ANOVA with Sidak's multiple comparison test.

FIGURE 4

The number of circulating TSPs and BM lymphoid progenitors declines between 1MO and 3MO of age. (a) Representative flow cytometry plots showing the percentage of Flk2⁺CD27⁺ progenitors within the Lin⁻ compartment of the blood in 1MO and 3MO mice. (b) Frequency and absolute cell numbers of circulating lymphoid progenitors in total blood of 1MO and 3MO mice. (c) Representative gating strategy for quantification of Lin⁻Flk2⁺CD27⁺ progenitors, Ly6d⁻ common lymphoid progenitors (CLPs), and Flk2⁺ multipotent progenitors (MPPs) in the BM. (d–f) Frequency and absolute numbers of (d) Flk2⁺CD27⁺ progenitors, (e) Flk2⁺ MPPs and (f) Ly6d⁻ CLPs in BM of 1MO and 3MO mice. Symbols represent data from individual mice at each age. Bars represent means ± SEM of data compiled from three independent experiments (n = 9 mice per age group). Statistical analysis was performed using Student t test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 5

Age‐associated changes in Notch signaling accompany the decline in BM lymphoid progenitors at 3MO. (a–c) Heatmaps show row‐normalized z‐scores of gene expression values indicating enrichment of (a) the Notch signaling pathway (Hallmark), (b) Notch target genes, and (c) the T‐cell differentiation (KEGG) pathway in 1MO versus 3MO Ly6d⁻ CLPs (n = 3 independent experiments). (d–g) Representative flow cytometry plots, as well as quantification of the frequencies and relative fluorescence intensities (RFIs) of Notch‐Venus reporter expression in (d, e) BM Flk2⁺ MPPs and (f, g) BM Ly6d⁻ CLPs. (h) Representative flow cytometry plots showing gating to distinguish Notch‐Venus low from Notch‐Venus high cells. (i) Frequencies of Notch‐Venus low and Notch‐Venus high Ly6d⁻ CLPs in 1MO and 3MO BM. (d–i) Data are pooled from five independent experiments (n = 9–12 mice per age group).

FIGURE 6

Age‐associated changes in Notch ligand expression and Notch signaling in the thymus correlate with the decline in ETPs at 3MO. (a) qPCR analysis showing relative Dll4 expression by CD45⁻ EpCAM⁺ Ly51⁺ cTECs from 1MO and 3MO C57BL/6J mice. Data are pooled from two biological experiments with 3–4 technical replicates per experiment, and expression levels were normalized to those at 1MO of age. (b, d) Representative flow cytometry plots and (c, e) quantification of the frequency of Dll4‐reporter⁺ cells and Dll4‐mCherry RFIs in (b, c) CD45⁻ EpCAM⁺ Ly51⁺ cTECs, and (d, e) CD45⁻ CD31⁺ endothelial cells (ECs) from Dll4‐mCherry reporter mice. Data are normalized to the average Dll4‐mCherry expression by (c) 1MO cTECs and (e) 1MO ECs. (c, e) Data are pooled from four independent experiments (n = 8–10 mice). (f) Representative histogram (left) and quantification of RFIs (right) of NOTCH1 cell surface expression on thymic ETPs. Data are normalized to the average NOTCH1 MFI of 1MO mice in each individual experiment. Data are pooled from three independent experiments (n = 7 mice per age group). (g) Representative flow cytometry plots and (h) frequencies and RFIs of Notch‐Venus reporter expression in thymic ETPs. Data are pooled from five independent experiments (n = 9–12 mice per age group). (c, e, f, h) Symbols represent data from individual mice at each age and bars represent means ± SEM. Statistical analysis was performed using Student t test, *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.