Longterm maintenance of human naive T cells through in situ homeostasis in lymphoid tissue sites

Abstract

Naïve T cells develop in the thymus and coordinate immune responses to new antigens; however, mechanisms for their long-term persistence over the human lifespan remain undefined. Here, we investigated human naïve T cell development and maintenance in primary and secondary lymphoid tissues obtained from individual organ donors aged 3 months-73 years. In the thymus, the frequency of double-positive thymocytes declined sharply in donors over age 40 coincident with reduced recent thymic emigrants (RTE) in lymphoid tissues, while naïve T cells were functionally maintained predominantly in lymph nodes (LN). Analysis of TCR clonal distribution by CDR3 sequencing of naïve CD4⁺ and CD8⁺ T cells in spleen and LNs reveal site-specific clonal expansions of naïve T cells from individuals >40 years of age with minimal clonal overlap between lymphoid tissues. We also identified biased naïve T cell clonal distribution within specific lymph nodes based on VJ usage. Together these results suggest prolonged maintenance of naïve T cells through in situ homeostasis and retention in lymphoid tissue.

Fig. 1. Alterations in Thymus structure and diminished thymopoiesis with age

(A) Representative histology of thymus sections from individuals of indicated ages visualized by H&E staining shown 10× magnification (top, scale bars of 200μm indicated at lower left), with an enlarged section shown below at 40× magnification (scale bars of 50μm at lower left). Hassall Corpuscle structures are denoted by red arrows in each field. (B) Average Number of Hassall Corpuscle structures per viewing pane in pediatric (0-2 years, n=8, white), and adult (n=6, aged 23,25,30,49,50,57 years, black) donors. P-value denoted by **=0.0002. (C) Left: Representative flow cytometry plots indicating the coordinate expression of CD4 and CD8 on T cells in thymus tissue from donors of indicated ages, delineating double positive (DP) and CD4 and CD8 single positive (SP) subsets with percent of each indicated in quadrants. Right: CD69 expression by DP and SP subsets from thymus tissue obtained from donors of indicated ages. (D) Compiled percentage of DP thymocytes in thymus tissue of donors of indicated ages and gender (blue, male; red, female) from 27 donors (See Table S1).

Fig. 2. Naïve CD4⁺ and CD8⁺T cells are differentially maintained in lymphoid sites

(A) Mean frequencies ±SEM of naïve (CCR7⁺CD45RA⁺) CD4⁺ (left) and CD8⁺ (right) T cells in indicated tissues from donors stratified into 3 age groups: pediatric (0-2 years, red, n=18), younger adults (15-40 years, black dashed, n=27), and older adults (41+ years, black, n=24) donors. See individual means, SEM and p values in Table S2. (B) Naïve CD4⁺ (top) or CD8⁺ (bottom) within four indicated lymphoid tissue sites as a function of age with each dot representing an individual donor compiled from 69 donors (see Table S1). Average CD8:CD4 ratios SP, 31:35, p=0.16; ILN, 57:42, p<0.001; LLN, 49:38, p<0.001; MLN, 48:41, p=0.01) (C) Naïve T cell frequency in multiple tissue of individual donors (total=40) where all four tissue sites were obtained, shown as a heat map of highest and lowest percent naïve T cells ranked by age in donors

Fig. 3. Naïve T cells retain functionality independent on the extent of recent thymic emigrants (RTE)

(A) RTE content among naïve T cells was measured by quantitative of T cell receptor excision circles (TREC) using a PCR-based approach (See online methods). Graphs show TREC content per 100,000 sorted CD4⁺ (top) and CD8⁺ (bottom) naïve T cells from indicated lymphoid sites (spleen, circle; ILN, upside down triangle; LLN, square; MLN, triangle), SP cells from each thymus (indicated by “X”) and total T cells from tissues obtained from the 17yr old and 73yr old donors. Each color represents an individual donor (individual values in Table S3). (B) Naïve and TEM CD4⁺ and CD8⁺ T cells were sorted from spleen, ILN and MLN, stimulated with anti CD3/CD28/CD2 beads, and the cytokine content in supernatants was assessed using the BD Cytokine Bead Array kit (see methods). Shown are IL-2 and IFN-γ production (pg/ml, mean± SEM) by naïve CD4 and CD8 T cells isolated from tissues of donors under age 35yrs of age (white bars, 2-4 donors) and over 50yrs of age (black bars, 2-4 donors except for spleen CD4 T cells which is from 1 donor). (C) IL-2 and IFN-γ production by naïve and memory (TEM) CD4 and CD8 T cells from the LLN of individual donors <35 years (4 donors: 29, 25, 26, and 34 years) and >50 years (4 donors: 54, 56, 52, and 59 years) of age. Cytokine levels are normalized by donor and indicated by Z score ((cytokine level-mean cytokine level)/standard deviation). Values for each cytokine measured are shown in Table S4.

Fig. 4. TCR Repertoire diversity of naïve T cells decreases with age and is distinct from TEM subsets

Naïve (CD45RA⁺CCR7⁺) CD4⁺ and CD8⁺ T cells were sorted from spleen, inguinal and mesenteric lymph nodes (ILN, MLN) and CDR3β sequences were amplified and sequenced (See methods). (A) Repertoire diversity within spleen (red), ILN (blue) and LLN (green) tissues is quantified in bulk by Simpson's index (see methods) and plotted against donor age, as shown for CD4⁺ (left) and CD8⁺ (right) lineages compiled from 11 and 13 donors, respectively (see Table S3), with each dot representing a single tissue from an individual donor. (B) Maximum clone frequency is higher for memory compared to naïve T cells. The frequency of the largest sequenced clone in the sample is plotted for CD4⁺ and CD8⁺ T cells for both Naïve and TEM subsets. Significant p-values based on students' t-test are indicated between CD4⁺ and CD8⁺ TEM cells, and between Naïve and TEM populations within the CD4⁺ or CD8⁺ lineage. (C) Clonal diversity of every observed VJ combination as computed by Shannon entropy for naïve and TEM subsets for CD4⁺ (top) and CD8⁺ (bottom) lineages. Naive populations are depicted by triangles, TEM populations by circles. Clones were pooled from every donor and separated by their VJ cassette. The curve log2N depicts the maximum possible diversity for a fixed number of clones.

Fig. 5. Naïve repertoire exhibits minimal sharing between tissues

The nucleotide sequence of each clone in multiple tissues from individual donors was analyzed for clonal overlap. (A) Venn diagrams displaying nucleotide sequence overlap of the top 1000 clones naïve and TEM CD4⁺ (top) and CD8⁺ (bottom) clones between three tissues sites (Spleen,red; ILN,blue; LLN,green) of a 21year old (left, Donor 125) and a 51year old (right, Donor 201) donor. Actual numbers are slightly greater than 1000 to account for additional clones present in identical read numbers to the one thousandth clone. (B) Overlap of naïve (gold) or TEM (Blue) clones between tissue pairs as a function of clone frequency (quantified as read count) for the representative donors shown in (A). For clones with a given read count in the first tissue, the fraction of overlap with all clones in the second tissue is plotted. Counts greater than 20 are logarithmically binned into 25 bins. (C) Overlap versus clone frequency plot calculated as in (b) where for every tissue clones have been pooled from multiple donors analyzed. The number of donors is indicated in each individual graph.

Fig. 6. VJ usage between tissues shows divergence of the naïve T cell repertoire among older donors

(A) The top 50 VJ frequencies for CD4+ T cells from Spleen and LLN are plotted side by side for two representative donors, age 21 and age 51. Corresponding VJ distance for the sample is given in the top right corner. (B) The VJ-distance for 8 donors (7 CD4⁺, 7 CD8⁺). Distance is computed for every pair of tissues from the same donors and plotted against donor age. Distance between spleen and lung lymph node (white), inguinal lymph node and spleen (gray), lung lymph node and inguinal lymph node (black).