Adaptation by naïve CD4+ T cells to self-antigen-dependent TCR signaling induces functional heterogeneity and tolerance

Abstract

Naïve CD4⁺ T cells experience weak T cell receptor (TCR) signals induced by self-peptides presented by MHC II. To investigate how these "basal" TCR signals influence responses to agonist TCR ligand stimulation, we analyzed naïve CD4⁺ cells expressing varying amounts of CD5, Ly6C, and Nur77-GFP, markers that reflect the strength of basal TCR signaling. Phenotypic analyses indicate that the broadest range of basal TCR signal strength can be visualized by a combination of Nur77-GFP and Ly6C. A range of basal TCR signaling is detectable even in populations that express identical TCRs. Whereas moderate basal TCR signal strength correlates with higher IL-2 secretion at early time points following TCR stimulation, weak basal TCR signaling correlated with higher IL-2 secretion at later time points. We identify a population of Nur77-GFP^HI Ly6C^- cells that could not be reliably marked by either of CD5, Ly6C, or Nur77-GFP alone. These cells experience the strongest basal TCR signaling, consistently produce less IL-2, and express PD-1 and markers associated with anergy, such as Grail and Cbl-b. We propose that adaptation to the strength of basal TCR signaling drives the phenotypic and functional heterogeneity of naïve CD4⁺ cells.

Keywords: CD5; Nur77; T cell activation; T cell anergy; basal TCR signaling.

Fig. 1.

Increased Nur77-GFP and CD5 inversely correlates with IL-2 secretion. (A) Overlaid histograms show the expression of GFP, CD69, and CD25 on the sorted GFP^LO (red), GFP^MED (blue), and GFP^HI (orange) populations that were not stimulated (Upper row) or stimulated (Lower row) before the IL-2 secretion assay. Non-TCR transgenic T cells were stimulated with APCs plus soluble anti-CD3 antibodies, OT-II TCR transgenic cells were stimulated with APCs plus OVA peptide, and AND TCR transgenic T cells were stimulated with I-E^k–expressing APCs from C3H mice, plus MCC peptide. (B) Contour plots show the expression of GFP and secreted IL-2 in the indicated unstimulated and stimulated samples. The numbers indicate the percentage of events within the IL-2⁺ gate. (C) Bar graph displays the normalized ratio of the percentages of IL-2⁺ cells from the GFP^LO sample divided by the GFP^HI sample. Symbols indicate the individual replicates and the bars represent the mean. Error bars represent SEM. Data shown are representative of n = 3 or 6 independent experiments. (D) Contour plots show the expression of CD5 and secreted IL-2 in the indicated unstimulated and stimulated samples. The numbers indicate the percentage of events within the IL-2⁺ gate. (E) Bar graph displays the normalized ratio of the percentages of IL-2⁺ cells from the CD5^LO sample divided by the CD5^HI sample. Symbols indicate the individual replicates and the bars represent the mean. Error bars represent SEM. Data shown are representative of n = 3 independent experiments.

Fig. 2.

Early IL-2 responses differ between CD5^HI and Nur77-GFP^HI cells. (A) Histogram shows cells sorted for low or high CD5 expression. Contour plots show CD5 and IL-2 staining after an IL-2 secretion assay. Numbers represent the percentage of cells within the indicated gate. Data are representative of n = 3 independent experiments. (B) Histogram shows cells sorted for low or high Nur77-GFP expression. Contour plots show GFP and IL-2 staining after an IL-2 secretion assay. Numbers represent the percentage of cells within the indicated gate. Data are representative of n = 3 independent experiments. (C) Bar graph shows the mean log₂ fold-change between the percentages of IL-2–secreting cells in the CD5^LO or GFP^LO population divided by the percentages of IL-2–secreting cells in the CD5^HI or GFP^HI population. Error bars represent SD. Data are from n = 3 independent experiments.

Fig. 3.

Heterogeneous expression of Nur77-GFP, CD5, and Ly6C on naïve CD4⁺ cells. (A) Contour plots show the expression of the indicated markers on CD4⁺ CD44^LO CD62L^HI Foxp3-RFP⁻ cells from the indicated mouse strains. (B) Contour plots show the expression of the indicated markers on total CD4⁺ Foxp3-RFP⁺ cells from the indicated mouse strains. (C) Contour plots show Nur77-GFP and Foxp3-RFP expression on total CD4⁺ cells from the indicated mouse strains. (D) Overlaid histograms show the expression of Nur77-GFP by Ly6C⁻ and Ly6C⁺ Tregs. Data shown are representative of n = 3 independent experiments.

Fig. 4.

Nur77-GFP^HI Ly6C⁻ cells are a population not consistently identified by a CD5^HI phenotype. (A) Schematic shows the 4 arbitrary gates based on Nur77-GFP fluorescence and Ly6C staining used in this study. Letters indicate the designation applied to each population. (B) Bar graph shows the percentages of CD5^LO or CD5^HI cells that fall within population A through D gates in the indicated mouse strains. (C) Contour plots show the expression of Nur77-GFP and Ly6C on the total naïve CD4⁺ cell population, CD5^LO and CD5^HI cells. CD5^LO are the lowest 15% and CD5^HI cells are the highest 15% of naïve CD4⁺ cells by CD5 expression. Numbers indicate the percentage of cells within each gate, and are the numbers presented in the graphs in B. Representative of n = 3 independent experiments. (D) Bar graphs show the percentages of IL-2–secreting cells from sorted populations A to D, after 4 or 16 h of stimulation with APCs and anti-CD3. Error bars represent SEM. Data are from n = 3 experiments.

Fig. 5.

Attenuated agonist-induced TCR signal strength in cells with high basal TCR signaling. (A) Contour plot shows GFP and Ly6C expression on non-TCR transgenic CD4⁺ CD44^LO cells, and the 4 gates indicating populations A to D. Overlaid histogram shows the ratio of (Ca²⁺-bound Indo-1) divided by (Ca²⁺-unbound Indo-1) over time for each of the 4 gated populations. Arrows indicate the time of addition of soluble anti-CD3 or cross-linking secondary antibodies. Data are representative of n = 3 independent experiments. (B) Contour plots show staining intensity of ERK phospho-Thr-202/Tyr-204 and Nur77-GFP in unstimulated and anti-CD3 stimulated non-TCR transgenic CD4⁺ cells. Overlaid histograms show the distribution of cells gated on low or high GFP expression and the intensity of phospho-ERK staining in the GFP^LO (red) and GFP^HI (blue) populations. Data representative of n = 3 experiments.

Fig. 6.

Decreased IL-2 response by Nur77-GFP^HI Ly6C⁻ cells is not rescued by bypassing TCR-proximal signals. (A) Contour plots show expression of CD5 and secreted IL-2 following stimulation with PMA and ionomycin for 16 h. (Left) Sorted CD5^LO cells and (Right) sorted CD5^HI cells from the indicated mouse strains. Numbers indicate the percentage of cells within the IL-2⁺ gate. Data representative of n = 2 or 3 experiments. (B) Graph displays the percentage of CD5^LO and CD5^HI staining positive for secreted IL-2 following PMA and ionomycin stimulation. Individual dots indicate replicates and the lines connect data points from the same experiment. (C) Contour plots show the expression of Nur77-GFP and secreted IL-2 by populations A to D following stimulation with PMA and ionomycin. Numbers indicate the percentage of cells within the IL-2⁺ gate. Data representative of n = 3 experiments. (D) Bar graph displays the percentage of IL-2 secreting cells following PMA and ionomycin stimulation. Bars represent the mean and error bars represent SEM, n = 3 individual experiments.

Fig. 7.

The stability of GFP and Ly6C expression on naïve CD4⁺ cells in vivo. (A, Upper) Contour and dot plots show GFP and Ly6C expression on sorted donor cells before adoptive transfer. Letters indicate the 4 cell populations sorted. (Lower) Contour plots show GFP and Ly6C expression by donor cells 4- and 10-d posttransfer into WT CD45.1⁺ hosts. Numbers indicate the percentage of events within each gate. Data representative of n = 2 experiments. (B) Contour plots show GFP and Ly6C expression on donor non-TCR transgenic Foxp3⁻ CD4⁺ CD44^LO cells pretransfer and posttransfer into WT or MHC class II-deficient hosts. Data representative of n = 2 experiments. (C) Naïve, Foxp3⁻ OT-II TCR transgenic CD4⁺ cells were sorted based on GFP and Ly6C expression into four populations (A to D as shown in dot plot). Sorted cells were adoptively transferred into congenic WT hosts, which were immunized with OVA peptide-pulsed APCs. Contour plots show GFP and Ly6C expression on donor cells 72 h after immunization. Histograms show dilution of CellTrace violet by the indicated populations. Data representative of n = 2 experiments.

Fig. 8.

Anergic phenotype in naïve cells with the highest basal TCR signal strength. (A, Top) Contour plots show the gating scheme for populations A to D based on GFP and Ly6C expression. Shown are naïve CD4⁺ CD44^LO CD62L^HI Foxp3⁻ cells on non-TCR transgenic, OT-II, or AND backgrounds. (Middle) Histograms show PD-1 expression on populations A to D. (Bottom) Contour plots show GFP and PD-1 expression on the indicated populations. Data representative of n = 3 experiments. (B) Bar graph shows the relative expression of Grail (Rnf128) mRNA expression from non-TCR transgenic cells sorted with population A to D phenotype based on GFP and Ly6C expression. Values were normalized to a sample of total naïve CD4⁺ cells. Error bars represent SD. Data representative of n = 2 independent experiments. (C) Immunoblot analysis of Cbl-b expression in whole cell lysates of naïve CD4⁺ CD44^LO cells sorted based on GFP and Ly6C (populations A to D). Numbers represent the ratio of Cbl-b to ERK 2, normalized to population A. Data representative of n = 2 experiments. (D) Naïve Foxp3⁻ OT-II transgenic cells were sorted based on GFP and Ly6C expression as shown in the pre- and postsort contour plots. Sorted cells were stimulated with APCs and OVA peptide with exogenous IL-2 alone or in combination with 10 ng/mL TGF-β. Contour plots show the expression of Foxp3-RFP after 72 h in culture. Numbers indicate the percentage of events within the indicated quadrant. Data are representative of n = 2 experiments.