The histone peptide H4 71-94 alone is more effective than a cocktail of peptide epitopes in controlling lupus: immunoregulatory mechanisms

Abstract

Tolerance therapy with nucleosomal histone peptides H4(71-94), H4(16-39), or H1'(22-42) controls disease in lupus-prone SNF1 mice. It would be clinically important to determine whether a cocktail of the above epitopes would be superior. Herein, we found that compared with cocktail peptides, H4(71-94) monotherapy more effectively delayed nephritis onset, prolonged lifespan, diminished immunoglobulin G autoantibody levels, reduced autoantigen-specific Th1 and Th17 responses and frequency of T(FH) cells in spleen and the helper ability of autoimmune T cells to B cells, by inducing potent CD8 Treg cells. H4(71-94) therapy was superior in "tolerance spreading," suppressing responses to other autoepitopes, nucleosomes, and ribonucleoprotein. We also developed an in vitro assay for therapeutic peptides (potentially in humans), which showed that H4(71-94), without exogenous transforming growth factor (TGF)-β, was efficient in inducing stable CD4(+)CD25(+)Foxp3(+) T cells by decreasing interleukin 6 and increasing TGF-β production by dendritic cells that induced ALK5-dependent Smad-3 phosphorylation (TGF-β signal) in target autoimmune CD4(+) T cells.

Fig. 1

Beneficial effect of low dose tolerance therapy using single or trio cocktail peptides. Incidence of severe lupus nephritis (a) and percent survival (b) of lupus-prone SNF1 mice injected with single (H4_71–94), trio (H1′_22–42, H4_16–39, and H4_71–94) peptides, or PBS (8 mice/group). c Both single and trio peptide therapies markedly reduced the levels of IgG autoantibodies in the serum of treated mice. d Single (H4_71–94) peptide therapy reduced the levels of IgG autoantibodies against RNP and RNA in the serum of treated mice, but not trio cocktail peptide therapy except anti-RNA IgG autoantibody. Sampling for (c) and (d), SNF1 mice were bled after one month of treatment (at 4 mo of age) and were assayed for levels of anti-anti-dsDNA, anti-ssDNA, anti-nucleosomes, and anti-histone IgG autoantibodies, or levels of anti-anti-RNA, anti-RNP, anti-RNA plus RNP, and anti-DNA/histone IgG autoantibodies. IgG (mean± SEM, OD at 405 nm, 5 mice/group). e With identical treatment regimens to those in (c), renal histologic features of lupus nephritis were evaluated. Single or trio peptide tolerance therapy lowered histopathology score of nephritis. f Representative histologic features of kidneys with identical treatment regimens to those in (c). Marked perivascular, interstitial, and glomerular infiltration of Th17 cells (arrow in the left panel) were detected only in the kidney of control-treated mice (original magnification, X200). a–b (log rank test): ⁺, P=0.0483; ^x, P=0.0414; *, P=0.0153; **P=0.00248. c–d: ⁺, P<0.05; ^x, P<0.02; *, P<0.01; **, P<0.001

Fig. 2

Single low-dose peptide therapy decreased both IFN-γ and IL-17 responses to relevant peptide epitopes and nucleosomes by lupus T cells in ELI-SPOT, but trio cocktail peptide therapy decreased IFN-γ and IL-17 responses to only lower concentration of those autoantigens. a–b T cells from PBS, H4_71–94, or trio peptide treated SNF1 mice were challenged with nucleosomes in various concentrations and then analyzed for Th1 and Th17 responses. c–d Splenic T cells from PBS or H4_71–94 treated SNF1 mice were challenged with another nucleosomal epitope (H4_16–39) and H4_71–94 peptide therapy decreased both Th1 and Th17 responses also to this epitope. e–f Splenic T cells from PBS or trio cocktail peptide treated SNF1 mice were also challenged with H4_16–39 peptide, but trio cocktail peptide therapy decreased Th1 and Th17 responses to only low concentration of H4_16–39 peptide. IFN-γ and IL-17 responses are expressed in mean±SEM positive spots per 1×10⁶ T cells from three experiments (five mice per group). Baseline IFN-γ and IL-17 spots in lupus T cells and APC cultures without Ag were less than 10 spots per 1 × 10^{6. x}, P<0.02; *, P<0.01; **, P<0.001

Fig. 3

Both single H4_71–94 peptide and trio cocktail peptide therapies can reduce T_FH cells. a ICOS⁺PD-1⁺CD4⁺ T_FH cells were reduced in spleens of single H4_71–94 and trio peptide treated mice. Splenocytes from single peptide, trio cocktail peptide- or PBS-treated mice were stained for ICOS and PD-1 in CD4 gated cells without any stimulation in vitro. b Representative dot plot of ICOS⁺PD-1⁺ T cells in CD4 gated population. c Both single H4_71–94 peptide and trio cocktail peptide therapies suppressed the increase of BCL-6 mRNA upon stimulation with nucleosomes in vitro. Splenocytes from H4_71–94–,trio cocktail peptide- or PBS-treated mice were stimulated with nucleosomes (20 μg/ml) in vitro for 18 h and then analyzed for fold increase of BCL-6 mRNA by real time PCR. ⁺, P<0.05; X, *, P< 0.01; **, P<0.001

Fig. 4

H4_71–94 single peptide and trio cocktail peptide therapies suppress anti-dsDNA (a), anti-ssDNA (b), anti-nucleosomes (c), and anti-histone (d) autoantibody production by T and B cells in the nucleosome stimulated helper assay. Baseline levels of IgG autoantibodies produced by B cells cultured by themselves were: anti-dsDNA, 0.01±0.005; anti-ssDNA, 0.05±0.006; anti-nucleosome, 0.02±0.001; anti-histone, 0.03±0.008 mg/dL. ⁺, P<0.05; ^x, P<0.02; *, P<0.01; **, P<0.001

Fig. 5

Therapy with H4_71–94 alone generates CD8⁺ Treg cells with stronger suppressive activity on autoreactive Th17 cells, but trio cocktail peptide therapy generates stronger CD4⁺CD25⁺ Treg suppressing Th1 autoreactivity. Suppressive activity of T cell subsets from treated mice were assessed on IFN-γ and IL-17 responses of unmanipulated SNF1 lupus T cells to nucleosomes presented by APC in the ELISPOT assay (ratio of Treg: lupus Th=1: 4). a CD4⁺CD25⁺ Treg cells from trio cocktail peptide therapy showed 2.5 fold higher suppressive activities on nucleosome-specific Th1 cells than CD4⁺CD25⁺ Treg cells from single H4_71–94 therapy. b CD8⁺ cells from trio cocktail peptide therapy also showed 1.3 fold higher suppressive activities on the Th1 cells at 1 μg/ml and 10 μg/ml nucleosome stimulation. c CD4⁺CD25⁺ Treg cells from either single or trio cocktail peptide therapy could not suppress nucleosome-specific Th17 responses, except for suppression of Th17 response by CD4⁺CD25⁺ Treg from H4_71–94 singly treated mice at 3 μg/ml nucleosome stimulation. d CD8⁺ T cells from trio cocktail peptide therapy showed significant suppressive activity on nucleosome-specific Th17 response at 0.3 μg/ml nucleosomes stimulation, but CD8⁺ Treg cells from H4_71–94 single peptide therapy showed higher suppressive activity at 10 μg/ml nucleosome stimulation than CD8⁺ Treg cells from trio cocktail peptide therapy. ⁺, P<0.05; ^x, P<0.02; *, P<0.01; **, P<0.001

Fig. 6

DCs from H4_71–94 single and trio cocktail peptide tolerized mice produced markedly increased amount of TGF-β on stimulation of nucleosomes (a) or CpG DNA (b). Amount of TGF-β in culture supernatants of DCs were measured by ELISA. DCs from H4_71–94 single peptide- and trio cocktail peptide-tolerized mice produced markedly decreased amount of IL-6 on stimulation of nucleosomes (c) or CpG DNA (d). Amount of IL-6 in culture supernatants of DCs were measured by ELISA. ⁺, P<0.05; ^x, P<0.02; *, P<0.01; **, P<0.001

Fig. 7

In vitro, DCs expand more H4_71–94 specific CD4⁺CD25⁺ Treg cells from whole T cells of H4_71–94-tolerized mice than those of trio cocktail peptide-tolerized mice. a Results from three experiments were compared by histogram (CFSE dilution) for proliferated CD4⁺CD25⁺Foxp3⁺ T cells. Numbers represent the mean percentage of three separate experiments (n=5). b Representative dot plot of CD25⁺Foxp3⁺ T cells in CD4⁺ T cells from experiment (a). c In vitro, H4_71–94 and trio cocktail peptide therapies induced CD4⁺CD25⁺Foxp3⁺ T cells in splenocytes of unmanipulated 3 month old SNF1 mice. Splenocytes were cultured in RPMI 1640 plus FBS and IL-2 (20U/ml) for 9 days in the presence of peptides, and then analyzed for CD25⁺Foxp3⁺ T cell in CD4 gated population as compared to that of culture with PBS (32.4%). d Representative dot plot of CD25⁺Foxp3⁺ T cells in CD4⁺ T cells from experiment (c). e Comparison of the induction of CD4⁺CD25⁺Foxp3⁺ T cells by peptide epitope to that by whole autoantigen (nucleosome) stimulations. CD90⁺ T cells and DCs from 3 month old unmanipulated SNF1 mice were cultured in the presence of various amounts of H4_71–94, trio cocktail peptides, and nucleosomes for 7 days. DC pulsed with H4_71–94 induced up to 4 folds higher levels of CD25⁺foxp3⁺ cells in CD4 gated cell population as compared to DC fed with nucleosomes at 1 μg/ml concentration. (F) Representative dot plot of CD25⁺foxp3⁺ T cells in CD4⁺ T cells. ⁺, P< 0.05; ^x, P<0.02; *, P<0.01; **, P<0.001

Fig. 8

Endogenous TGF-β dependent induction of LAP, Smad3 phosphorylation and expression of FoxP3 in CD4+ T cells by H4_71–94 or trio cocktail peptides in vitro. a Both H4_71–94 single and trio cocktail peptides (1 μg/ml) in vitro induced LAP⁺CD25⁺CD4⁺T cells in splenocytes of unmanipulated 3 month old SNF1 mice. Splenocytes from unmanipulated 3 month old SNF1 were cultured in the presence of 20U/ml IL-2 and peptides for 7 days. Results (on 3rd day) from three experiments were compared by histogram. b Representative dot plot of LAP⁺CD25⁺cells in CD4 gated population. Phosphorylation of Smad3 (p-Smad3) in CD4⁺ T cells by peptide. Splenocytes from 3 month old unmanipulated SNF1 mice were cultured in the presence of single H4_71–94 peptide or trio cocktail peptides (1 μg/ml). c Significant increase in the phosphorylation of Smad3 on day 1, day 2 and day 4. Culture with H4_71–94 peptide for 1 h showed highest increase of pSmad3 in CD4⁺ T cells. d Comparison of Smad3 phosphorylation by peptide at 1 h and 96 h time points. Both H4_71–94 and trio cocktail peptides increased pSmad3 in CD4⁺ T cells in vitro. e Representative dot plots of p-Smad3⁺CD4⁺ T cells in CD4 gated population. Quadrants are demarcated based on each group’s own isotype control staining of samples. Endogenous TGF-β is required in the induction of Foxp3⁺CD4⁺Treg cells in vitro. ALK-5 inhibitor (TGF-β signal inhibitor) blocked in vitro induction of Foxp3⁺CD25⁺CD4⁺Treg cells by H4_71–94 peptide in the culture of CD90⁺ T cells and DCs for 4 days (f) and in culture of splenocytes for 7 days (g). In (g), CD4⁺ T cell gated population contained 14.9% (cultures with PBS DMSO=dotted line), 26.5% (cultures with H4_71–94 DMSO=thin red line) and 13.6% (cultures with H4_71–94 and ALK-5=thick blue line) Foxp3⁺CD4⁺ Treg cells. ^x, P<0.02; *, P<0.01; **, P<0.001