Complement regulator CD46 temporally regulates cytokine production by conventional and unconventional T cells

Abstract

In this study we demonstrate a new form of immunoregulation: engagement on CD4(+) T cells of the complement regulator CD46 promoted the effector potential of T helper type 1 cells (T(H)1 cells), but as interleukin 2 (IL-2) accumulated, it switched cells toward a regulatory phenotype, attenuating IL-2 production via the transcriptional regulator ICER/CREM and upregulating IL-10 after interaction of the CD46 tail with the serine-threonine kinase SPAK. Activated CD4(+) T cells produced CD46 ligands, and blocking CD46 inhibited IL-10 production. Furthermore, CD4(+) T cells in rheumatoid arthritis failed to switch, consequently producing excessive interferon-gamma (IFN-gamma). Finally, gammadelta T cells, which rarely produce IL-10, expressed an alternative CD46 isoform and were unable to switch. Nonetheless, coengagement of T cell antigen receptor (TCR) gammadelta and CD46 suppressed effector cytokine production, establishing that CD46 uses distinct mechanisms to regulate different T cell subsets during an immune response.

Figure 1

IL-2 regulates T_H1 versus Tr1 effector function in CD4⁺ T cells activated with anti-CD3 and anti-CD46. (a) IFN-γ and IL-10 in supernatants of purified T cells not activated (NA) or activated for 72 h with various combinations of mAb to CD3 (α-CD3), mAb to CD28 (α-CD28) and mAb to CD46 (α-CD46; horizontal axis) plus IL-2 (25 U/ml). (b) IFN-γ and IL-10 in supernatants of purified T cells left unactivated (-) or activated for 72 h with various combinations of mAbs as in a plus various concentrations of IL-2 (wedge). (c) IL-2, IL-10 and IFN-γ in supernatants of purified T cells not activated or activated for 0–120 h (horizontal axis) with anti-CD3 plus anti-CD46 plus no exogenous IL-2 (IL-2 analysis) or IL-2 at 50 U/ml (IFN-γ or IL-10 analysis). (d) Cytokine secretion at 36 h by purified CD4⁺ T cells not activated or activated with various combinations of mAbs to CD3, CD28 and CD46 (above plots) plus IL-2 (dose, left margin) or neutralizing mAb to IL-2. Numbers in plots indicate percent IFN-γ⁺IL-10⁻ cells (top left), IFN-γ⁺IL-10⁺ cells (top right) or IFN-γ⁻IL-10⁺ cells (bottom right). (e) Proliferation of freshly purified CD4⁺ T cells mixed with supernatants of T cells that were activated for 36 h with anti-CD3 and anti-CD46 in the presence of IL-2 (50 U/ml), separated by flow cytometry based on their secretion profile (IFN-γ⁺IL-10⁻, IFN-γ⁺IL-10⁺ or IFN-γ⁻IL-10⁺) and cultured independently for further 18 h (low IL-2); the mixtures were activated with anti-CD3 plus anti-CD28 with neutralizing mAb to IL-10 (α-IL-10) or isotype-matched control antibody (isotype) and assessed at day 6 (middle and right). Far left (three bars), proliferation of freshly purified T cells activated with anti-CD3 and anti-CD28 in the presence of fresh media or supernatants of T cells activated for 72 h with anti-CD3 and anti-CD28. A₄₉₀, absorbance at 490 nm. *P < 0.05 and **P < 0.01 (Student’s t-test). Data are representative of eight (a) or three (c,e) experiments (mean ± s.d.), eight experiments with three independent donors (b) or six experiments (d).

Figure 2

CD46–IL-2 signals induce a switch from a T_H1 phenotype to a suppressive Tr1 phenotype in CD4⁺ T cells. (a) Flow cytometry of the induction of IL-10-secreting T cells by anti-CD46 from an initial T_H1 effector cell (1) or from a distinct cell subset (2). (b) T_H1 lineage induction in purified CD4⁺ T cells left not activated or activated with anti-CD3 and anti-CD46 plus IL-2 (50 U/ml IL-2) and isotype-matched control antibody or neutralizing mAb to IFN-γ or IL-10, presented as the percentage of cytokine-positive (Cyt⁺) cells: IFN-γ⁺IL-10⁻, IFN-γ⁺IL-10⁺ or IFN-γ⁻IL-10⁺ (key). *P < 0.05 and **P < 0.01 (Student’s t-test). Data are representative of three experiments (mean ± s.d.). (c) Secretion of IFN-γ and IL-10 by IFN-γ⁺IL-10⁻ cells isolated by flow cytometry from purified T cells activated for 36 h (primary stimulation (1°)) with anti-CD3 alone or with anti-CD3 and anti-CD46 in the presence of IL-2 (5 U/ml); the IFN-γ⁺IL-10⁻ populations were expanded for 4 d with IL-2 (5 U/ml), then restimulated (above plots; secondary stimulation (2°)) and assayed 18 h later. Numbers in plots indicate percent IFN-γ⁺IL-10⁻ cells (top left), IFN-γ⁺IL-10⁺ cells (top right) or IFN-γ⁻IL-10⁺ cells (bottom right). Data are representative of four experiments.

Figure 3

CD46-mediated signals contribute to the regulation of IL-2 expression by CD4⁺ T cells. (a) Secretion of IFN-γ, IL-10 and IL-2 by T cells either not activated or stimulated for 36 h with anti-CD3 and anti-CD46. Numbers in plots in top row indicate percent IFN-γ⁺IL-2⁻ cells (top left), IFN-γ⁺IL-2⁺ cells (top right) or IFN-γ⁻IL-2⁺ cells (bottom right); numbers in bottom row indicate percent IL-10⁺IL-2⁻ cells (top left), IL-10⁺IL-2⁺ cells (top right; also IFN-γ⁺ (data not shown)) or IL-10⁻IL-2⁺ cells (bottom right). (b) Intracellular flow cytometry analysis of ICER/CREM expression after 36 h of no activation or activation with various combinations of mAbs to CD3, CD28 or CD46 (key) or isotype-matched control antibody (Isotype control). (c) Immunoblot analysis of the nuclear translocation of ICER/CREM, assessed in cytoplasmic (C) and nuclear (N) protein fractions of purified T cells either not activated or activated for 24 or 48 h with anti-CD3 and anti-CD46 (3+46). Right margin, molecular size in kilodaltons (kDa). (d) Chromatin-immunoprecipitation analysis of the binding of ICER/CREM to the IL2 promoter in IFN-γ⁺IL-10⁻ cells (IFN-γ⁺) and IFN-γ⁻IL-10⁺ cells (IL-10⁺) induced with anti-CD3 and anti-CD46, presented as PCR amplification of an IL2 promoter–specific sequence from genomic DNA (control; left) or from DNA precipitated with anti-ICER/CREM (PP DNA; right). Right margin, size in base pairs (bp). Data are representative of three (a,b,d) or two (c) experiments.

Figure 4

The intracellular CYT-1 domain of CD46 and SPAK are required for IL-10 production in CD4⁺ T cells. (a) Secretion of IL-10 by Jurkat T cells stably transfected with the BC1 isoform (Jurkat-BC1) or BC2 isoform (Jurkat-BC2) of CD46 and either not activated or activated for 48 h with anti-CD3 and anti-CD46 or with anti-CD46 alone. (b) Immunoprecipitation (with a CD46-specific mAb or isotype-matched control mAb) of lysates of primary human CD4⁺ T cells activated for 15 min with various antibodies (above lanes), followed by immunoblot analysis with anti-CD46 (top) or anti-SPAK (bottom). Left margin: CD46, BC1 + BC2, CD46 isoform with BC region and either CYT-1 or CYT-2; CD46, C1 + C2, CD46 isoform with only the C region. Lysate (far left lane), untreated lysates from nonactivated cells (control). (c) IL-10 secretion by purified CD4⁺ T cells left untransfected (NT) or transfected with SPAK-specfic siRNA, nonspecific siRNA (Control) or buffer alone (Mock) and then either not activated or activated for 36 h with mAb to CD3 alone or mAb to CD3 and mAb to CD46 (key). (d) IL-10 production by Jurkat T cells left untransfected (Jurkat) or stably transfected with wild-type SPAK (Jurkat-SPAK) or a ‘kinase-dead’ version of SPAK (Jurkat-ΔSPAK) and either not activated or stimulated with anti-CD3 and anti-CD46 or with anti-CD46 alone. *P < 0.01 (Student’s t-test). Data are representative of three experiments (mean ± s.d. in a,c,d).

Figure 5

CD46 directly regulates γδ T cell function. (a) PCR analysis of CD46 isoform use by human γδ T cells (V_γ9V_δ2; middle) and CD4⁺ T cells (CD4; right) and in Chinese hamster ovary (CHO) cell lines transfected with single isoforms of CD46 (left). (b–d) Production of IL-10 (b) and IFN-γ and tumor necrosis factor (TNF; c) by human γδ T cells among PBMCs stimulated with HMBPP (10 nM) with (+) or without (-) mAb to CD46 in the presence (filled bars) or absence (open bars) of IL-2 (100 U/ml). (d) CD25 expression by V_γ9V_δ2 T cells stimulated with HMBPP (10 nM) or CD4⁺ T cells stimulated with mAb to CD3 plus mAb to CD28, with or without mAb to CD46 and in the presence or absence of IL-2 (as in b,c). MFI, mean fluorescence intensity. (e) Flow cytometry of purified cultures of CD4⁺ or V_γ9V_δ2 T cells (both >97% pure) activated with mAb to CD3 or HMBPP, respectively, in the presence of mAb to CD46 and IL-2 (100 U/ml). Numbers in plots indicate percent IFN-γ⁺IL-10⁻ cells (top left), IFN-γ⁺IL-10⁺ cells (top right) or IFN-γ⁻IL-10⁺ cells (bottom right). (f) Production of IFN-γ by purified V_γ9V_δ2 T cells with or without anti-CD46 treatment in the presence or absence of IL-2 (indicated as in b,c). D1–D8, donors 1–8. Data are representative of three experiments (a–d) or one experiment each with six (e) or two (f) different donors.

Figure 6

CD4⁺ T cells from patients with rheumatoid arthritis are defective in the IFN-γ–IL-10 switch induced by anti-CD3 and anti-CD46. (a) Frequency of cytokine-secreting cells among blood-derived T cells obtained from three healthy donors (HD-A–HD-C) and three patients with rheumatoid arthritis (RA-A–RA-C) and activated for 36 h with anti-CD3 and anti-CD46 plus IL-2 (50 U/ml), presented as the percentage of cytokine-positive cells (key; middle value of two data points). (b) IFN-γ and IL-10 in supernatants of T cells activated as described in a and then maintained in culture for 5 d with IL-2 (50 U/ml). (c) Frequency of cytokine-secreting cells among cells treated as described in b, then restimulated for 8 h with anti-CD3 and anti-CD46 plus IL-2 (50 U/ml); results presented as in a. (d) Frequency of cytokine-secreting cells among T cells from the synovial fluid of the left knee (JA-lk) and right knee (JA-rk) of a patient with juvenile arthritis and blood-derived T cells from two healthy donors (HD-D and HD-E) after primary stimulation with anti-CD3 and anti-CD46; results presented as in a. (e) Secretion of IFN-γ and IL-10 by the cells in d during population expansion as described in b. (f) Frequency of cytokine-secreting cells among the cells in d after secondary stimulation as in c. Data are representative of two experiments (a–f).

Figure 7

Engagement of CD46 by locally produced C3b drives IL-10 expression in CD4⁺ T cells stimulated with anti-CD3 and anti-CD28. (a) Flow cytometry of C3b surface deposition on purified CD4⁺ T cells either not activated or activated for 12 h with immobilized mAbs (key). Data are representative of four experiments. (b) Production of IL-10 by purified T cells either not activated or activated for 36 h with mAb to CD3 alone or mAb to CD3 plus mAb to CD28 (horizontal axis) plus media alone or media plus soluble CD46 (sCD46; key). NS, not significant; *P < 0.01 (Student’s t-test). Data are representative of three experiments.