Lck availability during thymic selection determines the recognition specificity of the T cell repertoire

Abstract

Thymic selection requires signaling by the protein tyrosine kinase Lck to generate T cells expressing αβ T cell antigen receptors (TCR). For reasons not understood, the thymus selects only αβTCR that are restricted by major histocompatibility complex (MHC)-encoded determinants. Here, we report that Lck proteins that were coreceptor associated promoted thymic selection of conventionally MHC-restricted TCR, but Lck proteins that were coreceptor free promoted thymic selection of MHC-independent TCR. Transgenic TCR with MHC-independent specificity for CD155 utilized coreceptor-free Lck to signal thymic selection in the absence of MHC, unlike any transgenic TCR previously described. Thus, the thymus can select either MHC-restricted or MHC-independent αβTCR depending on whether Lck is coreceptor associated or coreceptor free. We conclude that the intracellular state of Lck determines the specificity of thymic selection and that Lck association with coreceptor proteins during thymic selection is the mechanism by which MHC restriction is imposed on a randomly generated αβTCR repertoire.

Figure 1. Effect of coreceptor-free and coreceptor-associated Lck on thymic selection

(A) hCD2-driven transgenic constructs encoding wildtype (top) and C20A/C23A mutant (bottom) Lck proteins. (B) Abrogation of Lck binding to coreceptor proteins. Thymocyte lysates were precipitated with anti-CD4 or anti-CD8α and blotted for Lck. Numbers indicate total Lck band intensities relative to B6 (set equal to 1.0). Data represent 4 experiments. (C) CD5 levels on thymocytes relative to control B6 mice (mean ± SE, n=7 mice/group). (D and E) Numbers of γδ (D) and αβ (E) LN T cells (mean ± SE, n=9 mice/group). (F) Numbers above profiles indicate thymus cellularity (mean ± SE, n=10mice/group). (G) Ligand specificity of αβT cells in mixed lymphocyte reactions as assessed byproliferation-induced CFSE dye dilution on day 4. Numbers represent percentage of cellsthat underwent at least one division. Data represent 3 experiments. **** P<0.0001; *** P<0.001; ** P<0.01. See also Figures S1 and S2.

Figure 2. Lck sequestration and TCR signaling in the thymus

(A) CD5 on thymocytes relative to B6 (set at 100%). Mean ± SE, n=3 mice/group. (B) hCD2-driven transgenic constructs encoding wildtype (444) and tailless (44T) CD4 proteins named for their extra-cellular, transmembrane, and cytosolic domains. (C) Lck is sequestered away from TCR on thymocytes by the cytosolic tail of coreceptor proteins. Thymocyte lysates were precipitated by anti-CD4 or anti-TCRβ and blotted for Lck or TCRζ. Data represent 3 experiments. (D) In vitro TCR induced calcium mobilization. Thymocytes were loaded at 31^oC with Indo-1 and coated with biotinylated anti-TCR (5μg/ml) either alone or together with biotinylated anti-CD4 (1μg/ml). Crosslinking by avidin (black arrow) and addition of ionomycin (red arrow) are indicated. Data represent 2 experiments. (E) In vivo signaling of thymocytes. CD5 on thymocytes is shown relative tocontrol B6 mice (mean ± SE, n=10 mice/group). (F) In vivo signaling of thymocytes assessed biochemically. Lysates wereprecipitated with anti-TCRζ and blotted for phospho-zeta, ZAP70, or TCRζ. Datarepresent 2 experiments. **** P<0.0001; *** P<0.001.

Figure 3. Thymic selection of MHC-independent αβTCR in the absence of both MHC and coreceptors

(A) TCR expression on thymocytes from transgenic host mice (Quad^KORAG2^KOBcl-2^Tg). Black lines indicate TCR expression, grey lines indicate control staining. Numbers in panels indicate mean fluorescence intensity (MFI). Data represent 5 experiments. (B) Thymocyte profiles from host transgenic mice expressing transgenic TCR. Numbers in CD5 histograms indicate CD5 MFI. Numbers in CD69 and CCR7 histogramsindicate frequency of positive cells. Thymus cellularity is shown as mean ± SE (n=5 mice/group). (C) Numbers of αβ LN T cells in TCR transgenic mice (mean ± SE, n=5 mice/group). (D) Expression of CD62L and CD44 on αβT cells from TCR transgenic mice. Data represent 3 experiments. (E and F) Lymphopenia-induced proliferation triggered by A11 and B12A TCR is dependent on CD155 recognition. A11 and B12A αβT cells were CFSE labeled and injected (5×10⁶ cells/mouse) into CD155^+/+ or CD155^−/− lymphopenic hosts for 1 week. Numbers in histograms (E) represent the percentage of cells that underwent at least two divisions and summarized in (F). Mean ± SE of 3 mice/group. **** p<0.0001; * p<0.05. See also Figure S3.

Figure 4. Identification of CD155 as the thymic selecting ligand

(A) In vivo positive selection signaling by A11 and B12A TCR requires CD155. Numbers in CD5 histograms indicate MFI. Numbers in CD69 and CCR7 histograms indicate frequency of positive cells. Thymus cellularity is shown as mean ± SE (n=5 mice/group). (B) CD69 versus CCR7 profiles of A11 and B12A thymocytes. Numbers indicate frequency of mature CD69⁻CCR7^hi thymocytes (n=5 mice/group). (C) Numbers of αβ LN T cells (mean ± SE of 5 mice/group). (D) CD155 expression by electronically sorted thymic elements by quantitativePCR (left panel) and surface protein expression (right panel). Mean ± SE (n=3). (E) Thymic elements that induce CD155-specific thymic selection. Four groups ofmixed donor bone marrow chimeras were constructed with 1:1 mixtures of A11 andB12F donor cells and injected into irradiated host (Quad^KORAG2^KO) mice that expressedor lacked CD155. Numbers of A11 αβT cells in LN were determined after 8-10wks.Each circle represents one mouse (n=4). * p <0.05; **** p<0.0001; NS, not significant. See also Figure S4.

Figure 5. Positive selection signaling by MHC-independent TCR in coreceptor-sufficient mice lacking MHC

(A) Thymocyte profiles. Numbers in CD4 vs CD8α plots indicate cell frequencies. Numbers in CD5 histograms indicate MFI. Numbers in CD69 and CCR7 histograms indicate frequency of positive cells. Thymus cellularity is shown as mean ± SE (n=6 mice/group). (B) Numbers of αβ LN T cells (mean ± SE, n=6 mice/group). (C) Positive selection signaling occurs in DN thymocytes and is ligand specific. CD4 vs CD8α profiles (top) and CD5 and CCR7 expression on CD24^hiDN thymocytes (bottom). Data represent 4 experiments. (D) Early positive selection signaling is TCR specific. CD5 and CCR7 expression on gated DN and DP thymocytes. Data represent 4 experiments. (E) Early positive selection signaling is inhibited by early CD4 expression. CCR7 and CD24 expression on thymocytes from B12A and B12A.444 transgenic mice. Data represent 4 individual mice/group. See also Figure S5.

Figure 6. Effect on thymic selection of coreceptor-associated and coreceptor-free Lck

(A) Numbers of αβ LN T cells in coreceptor-deficient (left side) or coreceptor-sufficient (right side) mice (mean ± SE, n=3 mice/group). (B) Numbers of αβ LN T cells in kinase-deficient mice (mean ± SE, n=3 mice/group). ND, not done. (C) Profiles of thymocytes with endogenous Lck (black lines) or Lck^mut (colored lines) proteins. Numbers in CD4 vs CD8α plots indicate cell frequencies. Numbers in CD69 histograms indicate frequency of positive cells and is summarized in right panels (mean ± SE, n=3 mice/group). (D) Numbers of αβ LN T cells in TCR transgenic mice with endogenous Lck (black bar) or Lck^mut (gray bar) proteins. Reconstitution by Lck^mut proteins relative to endogenous Lck proteins (set at 100%) is displayed in the right panel (mean ± SE, n=3 mice/group). * p<0.05; **, p<.01; ***, p<0.001; ****, p<0.0001; NS, not significant.

Figure 7. Role of evolutionarily conserved CDR2β residues in thymic selection by an MHC-independent TCR

(A) Thymocyte profiles from MHC-deficient mice expressing A11 TCR whose TCRβ chain was wildtype or contained a point mutation in CDR2, either Y48A or E54A. Numbers in TCRβ and CD5 histograms indicate MFI. Numbers in CD69 and CCR7 histograms indicate frequency of positive cells. Numbers in CD69 vs CCR7 plots indicate frequency of mature cells. Thymus cellularity is shown as mean ± SE (n=4 mice/group). (B) Assessment of in vivo signaling of thymocytes. Unstimulated thymocyte lysates were precipitated with anti-TCRζ and blotted for phospho-zeta, ZAP70, or TCRζ. Data represent 2 experiments. (C) LN αβT cell numbers (mean ± SE, n=4 mice/group). * p<0.05; ****, p<0.0001.