A conserved CXXC motif in CD3epsilon is critical for T cell development and TCR signaling

Abstract

Virtually all T cell development and functions depend on its antigen receptor. The T cell receptor (TCR) is a multi-protein complex, comprised of a ligand binding module and a signal transmission module. The signal transmission module includes proteins from CD3 family (CD3epsilon, CD3delta, CD3gamma) as well as the zeta chain protein. The CD3 proteins have a short extracellular stalk connecting their Ig-like domains to their transmembrane regions. These stalks contain a highly evolutionarily conserved CXXC motif, whose function is unknown. To understand the function of these two conserved cysteines, we generated mice that lacked endogenous CD3epsilon but expressed a transgenic CD3epsilon molecule in which these cysteines were mutated to serines. Our results show that the mutated CD3epsilon could incorporate into the TCR complex and rescue surface TCR expression in CD3epsilon null mice. In the CD3epsilon mutant mice, all stages of T cell development and activation that are TCR-dependent were impaired, but not eliminated, including activation of mature naïve T cells with the MHCII presented superantigen, staphylococcal enterotoxin B, or with a strong TCR cross-linking antibody specific for either TCR-Cbeta or CD3epsilon. These results argue against a simple aggregation model for TCR signaling and suggest that the stalks of the CD3 proteins may be critical in transmitting part of the activation signal directly through the membrane.

Figure 1. Two cysteines in the stalk region of CD3ε are highly conserved and form a CXXC motif.

(A) Protein sequences of CD3ε from different species are aligned using the Clustal W program (http://www.ebi.ac.uk/clustalw/). The hypothetical boundaries of the stalk region are defined based on the mouse CD3ε protein. The two conserved cysteines in the stalk region are colored red and marked with the arrowheads. (B) A schematic view of constructed CD3ε proteins. Both of the non-mutated (ε^WT) and C-to-S mutated (ε^Mut) murine CD3ε primary sequences have a c-Myc and flash tag sequence on each end. cDNA fragments encoding these sequences were cloned into a human CD2 minigene cassette and subsequently generated transgenic mice were bred to endogenous cd3ε null background (see Materials and Methods for details).

Figure 2. Transgenic mice with cysteine mutations have defects in T cell development in the thymus.

(A) Total cell counts in various organs in the combined wt control (Cntl, open bars) versus the mutant transgenic mice (tgε^Mut, filled bars). Cell counts were obtained after the lysis of red blood cells. Data from lymph nodes represent cell counts from inguinal, brachial, and axillary lymph nodes combined and are shown as the number per lymph node. Data shown are for 8 tgε^Mut mice and 13 wt control mice (4 BL6, 1 tgε^WTcd3ε^+/−, 5 tgε^Mutcd3ε^+/−, and 3 tgε^WT). (B) Total cell counts in the thymus that were at the DN1, DN3, or DN4 stage. The stages are defined by CD44 and CD25 costaining, with a schematic view shown in the upper-left corner of the graph. Data shown are for 8 tgε^Mut mice (circles) and 13 wt control mice (squares) as in (A) and 2 cd3ε^−/− mice (diamonds). (C) Number of thymocytes that were at the double positive (DP) or single positive stages (CD4 SP and CD8 SP) in control (open bars) versus tgε^Mut (filled bars) mice. The numbers of mice examined are as in (A). (D) The percentage of CD5^hi cells at the double positive stage of T cell development in the thymuses of control (open bars) versus tgε^Mut (filled bars) mice. The numbers of mice examined are as in (A). (E) The percentage of CD69^hi cells at the double positive or single positive stages of T cell development in the thymuses of control (open bars) versus tgε^Mut (filled bars) mice. Data shown are for 4 tgε^Mut mice and 8 wt control (cntl) mice (3 BL6, 1 tgε^WTcd3ε^+/−, 1 tgε^Mutcd3ε^+/−, and 3 tgε^WT). For (B–E) cells from the thymus were costained with multiple antibodies against surface proteins (CD4, CD8, B220, CD44, CD25, CD5, and CD69). To avoid including non-T lineage cells and minimize nonspecific staining, only live B220⁻ cells were analyzed. The mean of the group and the SEM are indicated. p values were obtained from t tests. In some cases, fold decreases are indicated.

Figure 3. The numbers of mature T cells from transgenic mice bearing mutated CD3ε were decreased, and they were actively proliferating in a homeostatic manner.

In all panels data for control mice are open bars and for tgε^Mut mice are filled bars. (A) The cell counts of mature CD4⁺ or CD8⁺ T cells in the spleen. Fold decreases for the two populations are indicated. Data shown are for 8 tgε^Mut mice and 13 wt control mice (4 BL6, 1 tgε^WTcd3ε^+/−, 5 tgε^Mutcd3ε^+/−, and 3 tgε^WT). (B) The incorporation of BrdU into the mature T cells. Mice were given drinking water containing 0.8 mg/ml BrdU for 9 consecutive days before sacrifice. Data shown are three mice for each group. (C, D) The percentage of CD44^hiCD62L^lo or CD44^loCD62L^hi T cells in the CD4⁺ or CD8⁺ population of spleen and lymph nodes. The numbers of mice examined are as in (A). Spleen or pooled lymph node (inguinal, brachial, and axillary) cells were costained with antibodies against B220, CD4, CD8, CD44, CD62L, as well as BrdU in the case of (B). The percentages were obtained from flow cytometry data by gating on live B220⁻CD4⁺ or B220⁻CD8⁺ cells. Data are shown as the mean ± SEM. p values were obtained from t tests.

Figure 4. The cysteine mutations affected the CD4⁺ and CD8⁺ population unequally.

Spleen cells from control (4 BL6, 1 tgε^WTcd3ε^+/−, 5 tgε^Mutcd3ε^+/−, and 3 tgε^WT, open circles) or tgε^Mut (filled diamonds) mice were costained using anti-B220, anti-CD4, and anti-CD8 antibodies. The CD4 to CD8 ratio was calculated based on live B220⁻CD4⁺ or the B220⁻CD8⁺ cells. p values were obtained using t tests.

Figure 5. TCR signaling is compromised in the naïve peripheral T cells from tgε^Mut mice.

(A) ERK1/2 phosphorylation versus TCR occupancy in naïve CD4⁺ or naïve CD8⁺ T cells. As described in Materials and Methods, T cells from spleen and lymph nodes of tgε^WT (open circles) or tgε^Mut (filled circles) mice were nylon wool-purified and activated for 20 min with various concentrations of anti-Cβ antibody. The percentage of naïve (CD44^lo) CD4⁺ and CD8⁺ T cells containing pERK was determined. The data are presented as the percent of pERK⁺ cells versus the amount of anti-Cβ antibody bound (geometric mean fluorescence intensities, M.F.I.). The values of pERK⁺ were from the CD44^loCD4⁺ or CD44^loCD8⁺ populations, hence the “naïve CD4⁺” or “naïve CD8⁺”. In each experiment, cells of each group were pooled from two individual animals. (B) CD69 induction on naïve (CD44^lo) CD4⁺ or CD8⁺ T cells from tgε^WT (open bars) or tgε^Mut (filled bars) mice after T cell activation. Nylon wool-purified T cells were stimulated with nothing, plate-bound anti-TCR-Cβ (H57–597) antibody or 20 nM PMA for 12 h. Surface induction of CD69 was assessed by flow cytometry. Data are shown as the mean ± SD for three independent experiments. p values are from t test. (C) Naïve T cells from the tgε^Mut mice respond poorly to superantigen stimulation. Spleen and lymph node cells from tgε^WT (open bars) or tgε^Mut (filled bars) mice were cultured with SEB (10 µg/ml) for 15 h. The induction of CD69 on CD4⁺ and CD8⁺ naïve T cells bearing either Vβ8 (SEB-reactive) or Vβ6 (SEB-non-reactive) T cells was determined by flow cytometry. The data are expressed as the average increase in cells expressing CD69 (%) over that observed in control cultures that received no SEB. (D) The defect in signaling in the tgε^Mut T cells was intrinsic to the CD3ε molecule. Nylon wool-purified T cells from tgε^WT (open bars) or tgε^Mut (filled bars) mice were stimulated by plate-bound anti-TCR-Cβ or anti-CD3ε antibodies at 37°C for 20 min. After stimulation, CD4⁺ T cells were analyzed for intracellular pERK by flow cytometry. The dash line indicates the control cells that were not stimulated by the antibodies. Data (circles) are from two independent experiments. In each experiment, cells of each group were pooled from two individual animals.

Figure 6. Implications of effects of the CD3ε mutation on models of TCR signal transduction.

(A) Schematic representation of proposed models for signal transduction across the T cell membrane to initiate downstream signaling. For simplicity the ζ chain and CD4/CD8 co-receptors are omitted from the diagrams. “I” denotes the critical protein required for initiation of the signaling pathway. See text for discussion. (B) Predicted effect of the CD3ε mutation on the various models. See text for discussion.