CD6 regulates T-cell responses through activation-dependent recruitment of the positive regulator SLP-76

Abstract

Deciphering the role of lymphocyte membrane proteins depends on dissecting the role of a protein in the steady state and on engagement with its ligand. We show that expression of CD6 in T cells limits their responsiveness but that engagement by the physiological ligand CD166 gives costimulation. This costimulatory effect of CD6 is mediated through phosphorylation-dependent binding of a specific tyrosine residue, 662Y, in its cytoplasmic region to the adaptor SLP-76. A direct interaction between SLP-76 and CD6 was shown by binding both to a phosphorylated peptide (equilibrium dissociation constant [K(D)] = 0.5 muM at 37 degrees C) and, using a novel approach, to native phosphorylated CD6. Evidence that CD6 and SLP-76 interact in cells was obtained in coprecipitation experiments with normal human T cells. Analysis of human CD6 mutants in a murine T-cell hybridoma model showed that both costimulation by CD6 and the interaction between CD6 and SLP-76 were dependent on 662Y. The results have implications for regulation by CD6 and the related T-cell surface protein, CD5.

FIG. 1.

(A) Diagrammatic representation of CD5 and CD6, showing the three scavenger receptor cysteine-rich domains (rectangles) in their extracellular regions and the large cytoplasmic region of CD6. The membrane proximal domain of CD6 binds the N-terminal Ig superfamily domain (oval) of CD166. The SH2 domain of SLP-76 (black rectangle) is shown binding to the CD6 cytoplasmic region (Fig. 2). (B to G) Blocking of the CD166/CD6 interaction with CD6d3 mAbs. Three CD6d3 mAbs, OX126, OX125, and OX124, were tested for binding to CD6d3CD4d3 + 4-biotin immobilized on a BIAcore chip and for blocking of CD166CD4d3 + 4 (1 μM) binding. CD166CD4d3 + 4 (1 μM) was injected simultaneously over CD6d3CD4d3 + 4-biotin and CD4d3 + 4-biotin. (B) OX126 and OX126 Fab bound CD6d3 and blocked CD166 binding (C and D). (E) OX124 and OX125 bound CD6d3 and did not block CD166 binding individually (F) but did block it when used together (G). OX126 Fab fragments blocked the CD166/CD6 interaction, but the dissociation rate was too high to observe blocking effects in cellular experiments at the concentrations tested (data not shown). The bars indicate the period of injection. (H) Antigen (tetanus toxoid)-specific proliferation of human T cells was inhibited by OX126 CD6 mAb but not OX124 or OX125 CD6 mAb or an isotype nonbinding control (OX122, IgG1) mAb, all at 10 μg/ml.

FIG. 2.

The SH2 domain of SLP-76 binds specifically to the phosphorylated C-terminal CD6 peptide. (A) SLP-76 was isolated from Jurkat T-cell lysates in pulldown experiments using a peptide containing the phosphorylated C-terminal tyrosine of CD6, CD6(662)P. Coomassie blue-stained reducing gel shows that a single, specific band was isolated with CD6(662P) and not with a control CD6 peptide, CD6 (632). This band was shown to be SLP-76 by tryptic digestion and mass spectrometry (arrow). (B to D) SLP-76 SH2 (1.5 μM) or SHIP SH2 (19.3 μM), indicated by the bar above the trace, were injected (5 μl) over immobilized peptides at 37°C (amounts immobilized are given in response units [RU]). (B and C) CD6(662)P (CD6P, 220 RU); CD6 (662) (CD6, 231 RU); CD5(487)P (CD5P, 406 RU); and (streptavidin) control or FcRγIIb(292)P (FcRP, 244 RU). (D) CD6(662)P (CD6P, 341 RU); CD6 (662) (CD6, 329 RU); FcRγIIb(292)P (FcRP, 328 RU); and CD6 (632) (control, 488 RU). (E and F) Equilibrium binding data for a range of concentrations from the experiments for panels B and D are plotted, and the affinity for binding of SLP76-SH2 to CD6(662)P (CD6P) and SHIP-SH2 binding to FcRγIIb(292)P (FcR) were calculated from the fitted curve. Scatchard analyses of data are shown in panels G and H.

FIG. 3.

The SH2 domain of SLP-76 binds native CD6. Similar amounts of the CD6 mAb MEM-98 were immobilized (∼11,400 RU) on three BIAcore flow cells. Similar amounts of phosphorylated CD6 (CD6P; 1193 RU), phosphorylated 662F (662F;1664 RU), and unphosphorylated CD6 (CD6;1863 RU) were captured from hybridoma cell lysates. Increasing concentrations of SLP-76 were passed over CD6P, 662F, and CD6. (A) Data for two concentrations (μM) of the SLP-76 SH2 domain show binding to CD6P, weaker binding to 662F, and no binding to CD6, where only the bulk effect of the high-protein concentration is seen. (B) Specific equilibrium binding data for a range of concentrations from the experiment for panel A are plotted, and the affinity for binding of SLP-76SH2 to phosphorylated CD6 (CD6P) was calculated from the fitted curve. The affinity of SLP-76 SH2 for 662F was estimated by fixing the maximum binding at 19 RU to allow for there being more 662F on the chip (D) than CD6P (maximum, 14 RU). (C and D) The level of phosphorylation and amount of CD6 on the chip were quantitated by binding of phosphotyrosine and CD6 (OX124) mAbs. (E and F) Human CD6, CD2, CD5, or SLP-76 (SLP) were immunoprecipitated from Brij-96 lysates of human T blasts stimulated with CD3 mAb for 0, 2, or 5 min. Phosphorylated proteins (E) or SLP-76 (F) were detected by blotting with a pTyr mAb or a SLP-76 mAb, respectively. Arrows mark CD6 (∼130 kDa) and SLP-76 (∼68 kDa).

FIG. 4.

Human CD6 (hCD6) and mouse CD6 (mCD6) bind mouse CD166 with the same affinity, a K_D of ∼1 μM at 37°C. (A) Representative data for different concentrations (μM) of monomeric human CD6CD4d3 + 4 injected at 37°C over mouse CD166CD4d3 + 4-biotin, human CD166CD4d3 + 4-biotin (not shown), and a negative control, CD4d3 + 4-biotin, immobilized on a BIAcore chip. (B) Analysis of equilibrium binding from the experiment for panel A. (C and D) Data derived as for panels A and B for mouse CD6-CD4d3 + 4 injected over mouse CD166-CD4d3 + 4-biotin (C) and mouse CD166-CD4d3 + 4 over mouse CD6-CD4d3 + 4-biotin (D).

FIG. 5.

Costimulation of T-cell responses by CD6 is dependent on 662Y. (A) Human CD6 (thick line) and mutants, 662F (thin line), 489F (dotted line), and 489F/662F (dashed line) (CD6 mAb, T.12) are expressed at equivalent levels in 2B4 T-cell hybridoma cells and not on the parental cells (dashed-dotted line). Staining of CD6 mAb on untransduced cells was superimposable with a negative control (data not shown). Murine CD6 and CD3 relative to a negative control mAb (filled histogram) were expressed at similar levels on CD6 and mutant hybridoma cells. CD166 as detected by binding of mouse CD6 beads was highly expressed on APC and not on the T-cell hybridoma cells. FL1-H and FL2-H denote fluorescence intensity. (B and C) Antigen-specific IL-2 production by 2B4 hybridoma cells untransduced or expressing human CD6 or mutants in response to 1 μM mcc peptide and varying ratios of APC to T cells (B) and the APC-to-T-cell ratio of 10:1 with varying mcc peptide concentrations (C). (D) Mouse thymocytes were double labeled with phycoerythrin-anti TCRβ and CD6, CD5, or CD2 mAbs. CD6, CD5, and CD2 expression was increased on TCR-high (thick line) compared with TCR-low (thin line) thymocytes.

FIG. 6.

CD6 costimulation is dependent on CD166/CD6 and CD6/SLP-76 interactions. Antigen-specific (1 μM peptide) IL-2 production by 2B4 hybridoma cells at 24 h at the APC-to-T-cell ratio of 10:1, untransduced or expressing hCD6 or 662F in the presence of soluble hCD6-CD4d3 + 4 (sCD6; 6 μM) (A) or control soluble rat CD4d3 + 4 (srCD4; 6 μM) (B). (Upper panel). Human CD6 and mutants can be phosphorylated in mouse T-cell hybridoma cells. Human CD6 was immunoprecipitated from NP-40 lysates of 2B4 hybridoma cells expressing wild-type (WT) human CD6 and mutants or untransduced (2B4) treated or not with pervanadate (indicated by + or −) and analyzed by Western blotting with a pTyr mAb. (Lower panel) Human CD6 was detected in total lysate by blotting with CD6 mAb under nonreducing conditions. The lower band most likely represents incompletely processed human CD6. (C and D) Coprecipitation of CD6 and SLP-76 is dependent on 662Y. Human CD6 was immunoprecipitated with SLP-76 mAb or CD6 mAb as indicated above each lane from Brij-96 lysates of 2B4 hybridoma cells expressing human CD6 and mutants or untransduced 2B4 cells treated or not (first lane) with pervanadate and analyzed by Western blotting with a pTyr mAb (C) or a SLP-76 mAb (D). (E) Receptor-mediated phosphorylation of CD6 was dependent on 662Y. Human CD6 was immunoprecipitated from NP-40 lysates of 2B4 hybridoma cells transduced with wild-type human CD6 or 662F stimulated with CD3 mAb for 0 or 2 min and analyzed by Western blotting with a pTyr mAb.