Superior molecularly altered influenza virus hemagglutinin peptide 308–317 inhibits collagen-induced arthritis by inducing CD4+ Treg cell expansion

Authors


Abstract

Objective

To investigate the inhibitory effect and possible mechanism of a novel influenza virus hemagglutinin 308–317 peptide (altered HA308–317 peptide) in collagen-induced arthritis (CIA).

Methods

CIA was induced in DBA/1 mice by immunization with type II collagen (CII). Altered HA308–317 peptide, wild HA308–317 peptide, wild CII263–272 peptide, and irrelevant peptide were administered intranasally beginning at arthritis onset. Clinical and histologic scores were assessed, and cytokine levels were determined in the serum or in supernatants from splenocytes. Characteristics of T cell subsets in response to different peptides were analyzed both in vivo and in vitro.

Results

Intranasal administration of wild CII263–272 peptide, wild HA308–317 peptide, or altered HA308–317 peptide could significantly ameliorate CIA, but altered HA308–317 peptide showed greater therapeutic effects than wild CII263–272 peptide and wild HA308–317 peptide. The effect of altered HA308–317 peptide was associated with a substantial decrease in production of interleukin-17 (IL-17) and interferon-γ (IFNγ) and with a marked increase in production of IL-10 and transforming growth factor β, both in serum and in supernatants from splenocytes treated with altered HA308–317 peptide. Both the number and function of CD4+ Treg cells were significantly up-regulated by altered HA308–317 peptide, with a decreased induction of Th1 cells (CD4+IFNγ+) and Th17 cells (CD4+IL-17+). Adoptive transfer of CD4+CD25+ T cells from altered HA308–317 peptide–treated mice resulted in greater suppressive capacity in ameliorating CIA severity than did adoptive transfer of CD4+CD25+ T cells from wild HA308–317 peptide–treated, wild CII263–272 peptide–treated, or irrelevant peptide–treated mice.

Conclusion

Intranasal administration of altered HA308–317 peptide potently suppressed the severity of CIA by increasing the number and function of CD4+ Treg cells, suggesting that altered HA308–317 peptide might be a promising candidate for treatment of rheumatoid arthritis.

Rheumatoid arthritis (RA) is an autoimmune disease caused by loss of immunologic self-tolerance that leads to chronic inflammation in the joints and subsequent cartilage destruction and bone erosion. Many studies have demonstrated that the preferential binding of peptides by HLA–DR molecules and then activation of autoreactive T cells are involved in RA development (1, 2). In previous studies, Th1 cells were believed to be the main effector T cells activated by RA-specific antigens (3). However, this notion has been challenged by recent expeditious understanding of Th17 cells in autoimmunity (4). Th17 cells selectively produce interleukin-17 (IL-17) (5). The neutralization (6) or genetic deletion (7) of IL-17 in mice inhibits the development of collagen-induced arthritis (CIA). Moreover, Treg cells, which comprise 5–10% of CD4+ T cells and are marked by FoxP3 expression, are crucial for the maintenance of peripheral tolerance (8).

Evidence is accumulating that defects in the number or function of Treg cells are important in the immune imbalance that culminates in RA (9, 10). The balance between Th1/Th17 cells and Treg cells may play a pivotal role in determining the speed and severity of disease progression, at least in CIA (11). Therefore, increasing the number of Treg cells or enhancing Treg cell function may represent a therapeutic strategy for RA. Furthermore, the inflammatory cytokine milieu in the presence of transforming growth factor β (TGFβ) strongly induces either Th17 cell or Treg cell differentiation (12). The presence or absence of IL-6 is likely to be of particular importance, since stimulation of CD4+ T cells with IL-6 plus TGFβ potently induces Th17 cell differentiation, whereas stimulation with TGFβ alone triggers the development of Treg cells (12, 13).

Type II collagen (CII) has been implicated as one of the most important autoantigens involved in RA pathogenesis (14). CII256–271 peptide contains main antigenic epitopes that can trigger an RA-specific T cell response (15). The influenza virus hemagglutinin 308–317 peptide (HA308–317) (YVKQNTLKLA) shares a similar 3-dimensional structure with CII256–271 and can bind HLA–DR4/1 molecules with higher affinity (16). In our previous studies, we designed an altered HA308–317 peptide with amino acid substitutions at sites of T cell receptor (TCR) contact (YAKQATLALA). In vitro, the altered HA308–317 peptide inhibited T cell activation induced by wild HA308–317 peptide and wild CII263–272 peptide in the CII-specific T cell clones and peripheral lymphocytes of patients with RA (17, 18). However, its role in preventing or reversing chronic inflammation in mice with CIA has not been studied.

CIA is a well-established RA model and shares a number of clinical, histologic, and immunologic features with RA (19, 20). We induced CIA in DBA/1 mice whose I-Aq motif resembles most closely that of DR4 subtypes and could bind an RA-specific antigen peptide such as CII263–272 (21). In the present study, we reported our findings of the therapeutic effects of wild HA308–317 peptide and altered HA308–317 peptide in CIA and their capacity to influence the inflammatory response both in vivo and in vitro. Particularly, we demonstrated that altered HA308–317 peptide could effectively suppress CIA severity by up-regulating the number and function of Treg cells and suppressing the proliferation of Th1/Th17 cells, suggesting that altered HA308–317 peptide is a promising candidate for the treatment of RA.

MATERIALS AND METHODS

Peptide synthesis.

The wild CII263–272 peptide (FKGEQGPKGE), wild HA308–317 peptide (YVKQNTLKLA), altered HA308–317 peptide (YAKQATLALA), and irrelevant peptide (ALALTAQKAY) were synthesized by solid-phase techniques (SBS Genetech). The purity was >95%.

Animals.

Male DBA/1 mice, ages 6–8 weeks, were purchased from SLAC Laboratory Animal Center. They were housed under specific pathogen–free conditions and fed standard rodent chow and water ad libitum. The studies were approved by the Animal Care and Use Committee of Peking University People's Hospital.

Induction and clinical assessment of CIA.

Bovine CII (Chondrex) was emulsified in an equal volume of Freund's complete adjuvant (Sigma-Aldrich). Mice were immunized intradermally at the base of the tail with 100 μl of emulsion containing 150 μg CII. Three weeks later, the mice received a booster injection of 75 μg CII emulsified in Freund's incomplete adjuvant. Mice were monitored for signs of arthritis by 2 independent observers (JS, Y. Liu) under blinded conditions. Arthritis development was monitored using a macroscopic scoring system ranging from 0 to 4 per limb, yielding total scores of 0–16 per mouse, as detailed previously (22).

Peptide treatment protocols.

Treatment commenced after the onset of CIA (approximately day 26), when arthritis became well established (arthritis score >2). Two sets of independent experiments were performed. In each set, the sick mice were randomly divided into 4 groups (10 mice per group). Twenty-five micrograms of altered HA308–317 peptide, wild HA308–317 peptide, wild CII263–272 peptide, or irrelevant peptide dissolved in 10 μl phosphate buffered saline (PBS) was administered intranasally daily in each group until day 53. In the first set, the mice were killed on day 53. In the second set, arthritis severity was monitored until day 68 after the peptide treatment had been stopped for 15 days.

Histopathologic analysis of arthritis.

On day 53, the paws from each mouse were collected and then stained with hematoxylin and eosin. The histologic sections were analyzed microscopically by 2 observers (JS, Y. Liu) in a blinded manner, as detailed previously (23).

Cytokine detection.

Sera were collected on day 53, and the levels of cytokines (IL-10, TGFβ, IL-17, and IL-6) were measured by enzyme-linked immunosorbent assay (ELISA) kits (R&D Systems) according to the manufacturer's instructions.

Flow cytometric analysis.

The spleen cells were obtained and resuspended in RPMI 1640 medium. For intracellular cytokine analysis, cells were stimulated with 50 ng/ml phorbol myristate acetate plus 500 ng/ml ionomycin for 4–5 hours in the presence of 1 μg/ml brefeldin A (BD Biosciences). Cells were stained with fluorescein isothiocyanate (FITC)–conjugated anti-CD4 monoclonal antibody (mAb) (BD PharMingen) and then fixed and permeabilized, followed by intracytoplasmic staining using phycoerythrin (PE)–conjugated anti–interferon-γ (anti-IFNγ) or anti–IL-17 mAb (BD PharMingen). To determine the frequency of Treg cells, cells were labeled with FITC-conjugated anti-CD4 mAb and PE-conjugated anti-FoxP3 mAb. The percentage of cells staining positive was analyzed on a FACSCalibur flow cytometer using CellQuest software (Becton Dickinson).

In vitro assays.

On day 30 after immunization, mouse splenocytes were collected and incubated in RPMI 1640 at 37°C in 5% CO2 for 1 hour to remove adherent cells. Cells in the adherent cell fraction, which mainly consisted of macrophages, were used as antigen-presenting cells (APCs). Subsequently, CD4+CD25− T cells were isolated using an CD4+CD25+ Regulatory T Cell Isolation Kit (Miltenyi Biotec). Briefly, after the depletion of non-CD4+ T cells with a cocktail of biotinylated antibodies and Anti-Biotin MicroBeads (Miltenyi Biotec), the CD25+ MicroBeads were used for subsequent positive selection of CD4+CD25+ T cells. The rest of the cells (CD4+CD25−) were used as the responder T cells.

For proliferation assay, 2.5 × 105 purified CD4+CD25− T cells were cultured for 4 days in the presence of 1 × 105 irradiated splenic APCs per well in triplicate in 96-well plates and stimulated with 10 μg/ml wild CII263–272 peptide and different concentrations of wild CII263–272 peptide, wild HA308–317 peptide, altered HA308–317 peptide, or irrelevant peptide (0, 2, 10, and 50 μg/ml) in vitro. Cells were incubated at 37°C in 5% CO2 for 4 days. Before the last 16 hours of culture, 1 μCi of 3H-thymidine was added to each well. The cells were then harvested, and the incorporated radioactivity was counted. The 5,6-carboxyfluorescein succinimidyl ester (CFSE) method was also applied for proliferation assay. Briefly, purified CD4+CD25− T cells were resuspended in PBS containing 0.1% bovine serum albumin. CFSE (Invitrogen) was then added to the cell suspension to a final concentration of 10 μM. After incubation at 37°C for 15 minutes, the cells were washed extensively with RPMI 1640 medium. Cell samples were then cultured for 4 days in the system described above. The percentage of the cell population that was CFSE positive was measured by flow cytometry.

The production of IL-10, TGFβ, IFNγ, and IL-17 from the supernatants of isolated CD4+ T cells was determined by ELISA after 3 days of in vitro stimulation with peptides (10 μg/ml of wild CII263–272 peptide and 10 μg/ml of wild CII263–272 peptide, altered HA308–317 peptide, wild HA308–317 peptide, or irrelevant peptide). The frequencies of Th1, Th17, and CD4+FoxP3+ T cells were determined as described above.

Assessment of CD4+CD25+ T cell activity.

Splenocytes were harvested on day 53. CD4+CD25+ and CD4+CD25− cells were isolated as described above. The purity was confirmed to be >90%. The CD4+CD25+ cells (5 × 104 cells/well) from different peptide-treated mice with CIA were cultured with an equal number (1:1) of T responder cells (CD4+CD25−) from mice with CIA along with APCs (2 × 105 cells/well) and 10 μg/ml wild CII263–272 peptide for 4 days in RPMI 1640 containing 100 IU/ml IL-2 (R&D Systems). Thereafter, cells were pulsed with 1 μCi/well of 3H-thymidine for an additional 16 hours before harvesting. Radioactivity was counted. To evaluate the contribution of IL-10 and TGFβ to suppressive function of CD4+CD25+ T cells, neutralizing antibodies against mouse IL-10 (goat IgG), TGFβ (murine IgG1), or respective isotype control antibodies were applied to the above culture system, in which CD4+CD25+ T cells were derived from altered HA308–317 peptide–treated mice.

Adoptive transfer of CD4+CD25+ T cells from mice with CIA treated with different peptides.

On day 53, splenocytes were collected from mice with CIA treated with different peptides. Isolated CD4+CD25+ T cells (1 × 106) were resuspended in RPMI 1640 medium and injected intravenously into mice with established CIA. Arthritis development was monitored in recipient mice. Upon termination of the experiments, levels of cytokines (IL-6, tumor necrosis factor α [TNFα], IL-17, IL-10, and TGFβ) in serum were measured.

Statistical analysis.

Data analyses were performed using SPSS software, version 16.0. Results are expressed as the mean ± SD. Statistical differences were analyzed using Student's t-test or analysis of variance with Bonferroni adjustment. P values less than 0.05 were considered significant.

RESULTS

Treatment with altered HA 308–317 peptide has greater suppressive effects on CIA.

Administration of altered HA308–317 peptide significantly suppressed arthritis severity compared with administration of irrelevant peptide. As shown in Figure 1A, the difference between these 2 groups started on day 29 and continued until day 53. On day 53, the mean ± SD arthritis score in altered HA308–317 peptide–treated mice with CIA was much lower than that in irrelevant peptide–treated mice (2.62 ± 0.80 versus 7.73 ± 1.28; P < 0.01). Moreover, such therapeutic effects of altered HA308–317 peptide were still evident even after the peptide treatment had been stopped for 15 days (on day 68) (mean ± SD 2.65 ± 0.90 versus 7.45 ± 0.96; P < 0.01), indicating that a lasting immunosuppressive mechanism may have been induced (Figure 1B). The therapeutic effect of altered HA308–317 peptide on mice with CIA was further verified by histologic examination. Mice with CIA treated with irrelevant peptide developed characteristic chronic inflammation of synovial tissue, pannus formation, cartilage destruction, and bone erosion. In contrast, there was a remarkable reduction in inflammatory cell infiltration and pannus formation within the joint space of altered HA308–317 peptide–treated mice. Cartilage destruction and bone erosion were significantly ameliorated. The mean ± SD histologic score was markedly reduced in altered HA308–317 peptide–treated mice compared with irrelevant peptide–treated mice (0.55 ± 0.20 versus 2.03 ± 0.40; P < 0.01) (Figures 1C and D).

Figure 1.

Intranasal treatment with altered influenza virus hemagglutinin 308–317 peptide (altered HA308–317 peptide [aHAP]) strikingly ameliorates collagen-induced arthritis (CIA). After the onset of arthritis, mice with CIA were treated intranasally with altered HA308–317 peptide, wild HA308–317 peptide (wHAP), wild type II collagen (CII) 263–272 peptide (wild CII263–272 peptide [wCIIP]), or irrelevant peptide (IP) daily for 27 days. A and B, The mice were killed at the end of the treatment (on day 53) (A) or on day 68 after treatment had been stopped for 15 days (B) in 2 sets of independent experiments. Development of arthritis was monitored and is shown as mean ± SD arthritis scores. Analysis of variance with Bonferroni adjustment was used for multiple comparisons. C and D, At the end of the treatment, the paws from each mouse in the first set were stained and scored for histologic changes. Shown are quantitative histologic scores (C) and representative images (D). Results in C are expressed as the mean ± SD (n = 10 mice per group per experiment). ∗ = P < 0.05; ∗∗ = P < 0.01 versus irrelevant peptide. Δ = P < 0.05; ΔΔ = P < 0.01 versus wild HA308–317 peptide or wild CII263–272 peptide. Original magnification × 40.

Compared to irrelevant peptide treatment (mean ± SD clinical score 7.73 ± 1.28, histologic score 2.03 ± 0.40), treatments with wild HA308–317 peptide and wild CII263–272 peptide also resulted in reductions of clinical scores (5.11 ± 0.87 and 5.48 ± 1.02, respectively; both P < 0.01) and histologic scores (1.25 ± 0.20 and 1.27 ± 0.28, respectively; both P < 0.01) on day 53. Further, such effects were also evident after the peptide therapy had been stopped for 15 days. However, altered HA308–317 peptide conferred much better protection against CIA compared with wild HA308–317 peptide and wild CII263–272 peptide (both P < 0.01).

Significantly increased levels of IL-10 and TGFβ as well as reduced levels of IL-17 and IL-6 in serum from mice with CIA treated with altered HA 308–317 peptide.

To investigate the mechanisms that could be responsible for the effect of altered HA308–317 peptide, we examined serum levels of cytokines in mice with CIA treated with different peptides. Strikingly, the mean ± SD serum levels of both IL-10 and TGFβ in altered HA308–317 peptide–treated mice were substantially elevated (77.89 ± 16.58 pg/ml and 136.48 ± 6.58 pg/ml, respectively) compared with those detected in wild HA308–317 peptide–treated mice (14.97 ± 2.53 pg/ml and 105.57 ± 6.08 pg/ml, respectively; both P < 0.01), wild CII263–272 peptide–treated mice (14.67 ± 2.73 pg/ml and 103.64 ± 6.87 pg/ml, respectively; both P < 0.01), or irrelevant peptide–treated mice (6.61 ± 1.92 pg/ml and 66.57 ± 5.67 pg/ml, respectively; both P < 0.01) (Figure 2).

Figure 2.

Effects of intranasal altered HA308–317 peptide treatment on cytokine production. At the end of the treatment (on day 53), serum was collected from different groups of mice with CIA treated intranasally with irrelevant peptide, wild CII263–272 peptide, wild HA308–317 peptide, or altered HA308–317 peptide, and samples were analyzed for concentrations of interleukin-10 (IL-10), transforming growth factor β (TGFβ), IL-6, and IL-17. Results are expressed as the mean ± SD of 3 separate experiments (10 mice per group per experiment). ∗∗ = P < 0.01 versus irrelevant peptide. ΔΔ = P < 0.01 versus wild HA308–317 peptide or wild CII263–272 peptide. See Figure 1 for other definitions.

In contrast, the serum contained significantly lower levels of IL-6 (mean ± SD 26.99 ± 5.35 pg/ml) and IL-17 (14.99 ± 2.22 pg/ml) in altered HA308–317 peptide–treated mice, compared with those detected in wild HA308–317 peptide–treated mice (68.12 ± 13.01 pg/ml and 40.58 ± 3.45 pg/ml, respectively; both P < 0.01), wild CII263–272 peptide–treated mice (69.36 ± 13.66 pg/ml and 41.35 ± 4.69 pg/ml, respectively; both P < 0.01), or irrelevant peptide–treated mice (125.72 ± 14.31 pg/ml and 71.70 ± 3.27 pg/ml, respectively; both P < 0.01) (Figure 2). Although wild HA308–317 peptide and wild CII263–272 peptide also increased the levels of IL-10 and TGFβ and down-regulated IL-6 and IL-17 compared with irrelevant peptide (P < 0.01), such effects were much milder than those achieved with altered HA308–317 peptide (P < 0.01 versus wild HA308–317 peptide and P < 0.01 versus wild CII263–272 peptide).

Treatment with altered HA 308–317 peptide increases Treg cells but reduces Th1 and Th17 cell frequencies in vivo.

To further investigate the immunomodulating effect of altered HA308–317 peptide on T cell priming and differentiation, T cell subsets from mice with CIA treated with different peptides were examined. The CD4+FoxP3+ T cells (mean ± SD 8.11 ± 0.76%) from naive mice were considered the starting Treg cell population. We found that the proportion of CD4+FoxP3+ T cells increased in altered HA308–317 peptide–treated mice with CIA in vivo (33.10 ± 2.96%) compared with that in wild HA308–317 peptide–treated mice with CIA (24.09 ± 1.87%) (P < 0.01), wild CII263–272 peptide–treated mice with CIA (23.58 ± 1.82%) (P < 0.01), or irrelevant peptide–treated mice with CIA (16.52 ± 0.65%) (P < 0.01) (Figure 3A). In addition, altered HA308–317 peptide significantly decreased the frequency of IFNγ-producing Th1 cells (2.38 ± 0.75%) and IL-17–producing Th17 cells (2.61 ± 0.80%) compared with wild HA308–317 peptide (4.93 ± 1.02% and 4.67 ± 1.75%, respectively; both P < 0.01), wild CII263–272 peptide (5.24 ± 0.95% and 4.72 ± 1.22%, respectively; both P < 0.01), or irrelevant peptide (13.71 ± 2.62% and 8.31 ± 1.93%, respectively; both P < 0.01) (Figures 3B and C).

Figure 3.

Intranasal treatment with altered HA308–317 peptide increases frequency of CD4+ Treg cells and reduces Th1/Th17 cells in mice with CIA. AC, At the end of the treatment (on day 53), the percentages of CD4+FoxP3+ Treg cells (A), Th1 cells (CD4+ interferon-γ[IFNγ]+ T cells) (B), and Th17 cells (CD4+ interleukin-17[IL-17]+ T cells) (C) in splenocytes from mice in different groups were determined by flow cytometry. Results are expressed as the mean ± SD of 3 separate experiments (10 mice per group per experiment). ∗∗ = P < 0.01 versus irrelevant peptide–treated mice. ΔΔ = P < 0.01 versus wild HA308–317 peptide–treated mice or wild CII263–272 peptide–treated mice. See Figure 1 for other definitions.

Compared with treatment with irrelevant peptide, treatment with wild HA308–317 peptide and treatment with wild CII263–272 peptide also increased the frequency of CD4+FoxP3+ T cells (both P < 0.01), and the frequencies of Th1 cells and Th17 cells were also decreased (both P < 0.01), although the effects were much milder than those achieved with altered HA308–317 peptide (P < 0.01 versus wild HA308–317 peptide and P < 0.01 versus wild CII263–272 peptide).

Opposite effects of wild HA308–317 peptide and altered HA308–317 peptide on CII-primed T cell responses in vitro.

To determine whether altered HA308–317 peptide could suppress CII-specific T cell responses in vitro, we analyzed proliferation of CD4+CD25− T cells isolated from mice with CIA in response to different doses of peptides. As shown in Figure 4A, there was no difference between stimulation with 0, 2, 10, and 50 μg/ml of irrelevant peptide. However, compared with 0 μg/ml, the peptide-stimulated CD4+CD25− T cells showed stronger proliferative ability in the wild CII263–272 peptide and wild HA308–317 peptide groups at doses of 10 μg/ml and 50 μg/ml (P < 0.01) (data not shown). Furthermore, altered HA308–317 peptide–stimulated CD4+CD25− T cells at doses of 10 μg/ml and 50 μg/ml showed much weaker proliferative ability compared with 0 μg/ml (P < 0.01) (data not shown). However, different peptides at a dose of 50 μg/ml did not show any stronger effects than those same peptides at a dose of 10 μg/ml (P > 0.05). Therefore, we used peptides at 10 μg/ml in subsequent in vitro experiments. Compared with irrelevant peptide treatment at a dose of 10 μg/ml (mean ± SD 7,330.31 ± 705.89 counts per minute), wild HA308–317 peptide–stimulated and wild CII263–272 peptide–stimulated CD4+CD25− T cells obtained significantly stronger proliferative ability (9,147.00 ± 590.09 cpm and 9,063.67 ± 558.75 cpm, respectively; both P < 0.01). However, altered HA308–317 peptide inhibited CD4+CD25− T cell proliferation in comparison to irrelevant peptide (4,619.33 ± 539.46 cpm versus 7,330.31 ± 705.89 cpm; P < 0.01).

Figure 4.

Altered HA308–317 peptide–regulated T cell response and cytokine production in vitro. On day 30 after immunization, splenic CD4+ and CD4+CD25− T cells were isolated. A, The proliferation of CD4+CD25− T cells was measured by 3H-thymidine incorporation and labeling with 5,6-carboxyfluorescein succinimidyl ester (CFSE). The labeled CD4+CD25− T cells were cocultured with antigen-presenting cells (APCs) and stimulated with 10 μg/ml wild CII263–272 peptide and serial doses (0, 2, 10, and 50 μg/ml) of peptides as indicated. Proliferative activity was expressed as the mean ± SD cpm and the percentage of the population that was CFSE positive. ∗ = P < 0.05; ∗∗ = P < 0.01 versus 0 μg/ml of the same peptide. P < 0.01 for the other 3 peptides versus the irrelevant peptide at 10 μg/ml. B, Purified CD4+ T cells were cultured with APCs and then stimulated with peptides. The production of interleukin-10 (IL-10), transforming growth factor β (TGFβ), interferon-γ (IFNγ), and IL-17 from the supernatants of isolated CD4+ T cells was determined by enzyme-linked immunosorbent assay after 3 days of in vitro stimulation with peptides (10 μg/ml of wild CII263–272 peptide and 10 μg/ml of wild CII263–272 peptide, altered HA308–317 peptide, wild HA308–317 peptide, or irrelevant peptide). C, The frequencies of Treg cells (FoxP3+), Th1 cells (IFNγ+), and Th17 cells (IL-17+) in the cultures were determined. Results in B and C are expressed as the mean ± SD of 3 separate experiments (10 mice per group per experiment). ∗∗ = P < 0.01 versus T cells stimulated with irrelevant peptide. ΔΔ = P < 0.01 versus T cells stimulated with wild HA308–317 peptide or wild CII263–272 peptide. See Figure 1 for other definitions.

Additionally, there was no difference between wild HA308–317 peptide and wild CII263–272 peptide in terms of CD4+CD25− T cell proliferation (P > 0.05). The similar results were confirmed in the CFSE-labeled CD4+CD25− T cell proliferation assay. A smaller proportion of CD4+CD25− T cells remained undivided after 4 days of wild HA308–317 peptide treatment or wild CII263–272 peptide treatment compared with irrelevant peptide treatment. Furthermore, the CFSE dilution profiles showed striking differences in CD4+CD25− T cell proliferation between the irrelevant peptide–treated group and the altered HA308–317 peptide–treated group (Figure 4A and data not shown).

In addition, altered HA308–317 peptide treatment led to a significant reduction in IFNγ (mean ± SD 80.12 ± 16.01 pg/ml) and IL-17 (43.83 ± 24.13 pg/ml) production by sorted CD4+ T cells compared with wild HA308–317 peptide treatment (213.40 ± 37.19 pg/ml and 260.19 ± 48.34 pg/ml, respectively; both P < 0.01), wild CII263–272 peptide treatment (219.53 ± 31.09 pg/ml and 264.20 ± 42.21 pg/ml, respectively; both P < 0.01), or irrelevant peptide treatment (164.83 ± 24.39 pg/ml and 150.42 ± 18.78 pg/ml, respectively; both P < 0.01), whereas the production of the regulatory cytokines IL-10 (220.35 ± 59.99 pg/ml) and TGFβ (130.18 ± 26.72 pg/ml) was increased by altered HA308–317 peptide treatment compared with wild HA308–317 peptide treatment (71.56 ± 15.60 pg/ml and 66.46 ± 15.46 pg/ml, respectively; both P < 0.01), wild CII263–272 peptide treatment (68.33 ± 12.51 pg/ml and 62.73 ± 11.59 pg/ml, respectively; both P < 0.01), or irrelevant peptide treatment (91.50 ± 22.71 pg/ml and 76.13 ± 18.30 pg/ml, respectively; both P < 0.01). In contrast, compared with irrelevant peptide, both wild HA308–317 peptide and wild CII263–272 peptide up-regulated the production of IFNγ and IL-17 (both P < 0.01) and had no effect on the production of IL-10 and TGFβ (both P > 0.05) (Figure 4B).

Next, we determined the frequency of T cell subsets on sorted CD4+ T cells. As shown in Figure 4C, the frequency of CD4+FoxP3+ T cells was up-regulated in isolated CD4+ T cells stimulated with altered HA308–317 peptide (mean ± SD 20.05 ± 2.07%) in vitro compared with irrelevant peptide (9.35 ± 1.43%) (P < 0.01), wild HA308–317 peptide (5.01 ± 1.22%) (P < 0.01), or wild CII263–272 peptide (4.98 ± 1.09%) (P < 0.01). In addition, altered HA308–317 peptide treatment decreased the frequency of Th1 cells (5.42 ± 0.82%) and Th17 cells (4.33 ± 0.96%) compared to wild HA308–317 peptide treatment (13.08 ± 1.08% and 13.91 ± 1.57%, respectively; both P < 0.01), wild CII263–272 peptide treatment (12.93 ± 0.87% and 13.08 ± 0.79%, respectively; both P < 0.01), or irrelevant peptide treatment (8.79 ± 0.90% and 10.11 ± 0.81%, respectively; both P < 0.01). The results suggested that altered HA308–317 peptide partially inhibited the differentiation of autoreactive/inflammatory Th1 and Th17 cells, a finding that was in accordance with the results of the production of IFNγ and IL-17 in supernatants. However, compared with irrelevant peptide stimulation, wild HA308–317 peptide stimulation and wild CII263–272 peptide stimulation decreased the frequency of CD4+FoxP3+ T cells and up- regulated Th1 and Th17 cell subsets in vitro (all P < 0.01).

Greater in vitro suppressive capacity of CD4+CD25+ Treg cells from altered HA308–317 peptide– treated mice with CIA via a mechanism involving IL-10 and TGFβ.

The total numbers of Treg cells increased in spleen cells from altered HA308–317 peptide–treated mice with CIA. Thus, we addressed the question of whether CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice could suppress effector T cell responses in vitro. CD4+CD25− T cells from mice with CIA were cocultured with APCs stimulated with wild CII263–272 peptide in the presence or absence of CD4+CD25+ Treg cells isolated from mice with CIA treated intranasally with different peptides. As shown in Figure 5A, CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice with CIA were more effective at suppressing proliferation of CD4+CD25− T cells (mean ± SD 3,328.91 ± 701.32 cpm) than were CD4+CD25+ Treg cells from wild HA308–317 peptide–treated mice (5,484.43 ± 1,024.82 cpm) (P < 0.01), CD4+CD25+ Treg cells from wild CII263–272 peptide–treated mice (5,291.13 ± 799.35 cpm) (P < 0.01), or CD4+CD25+ Treg cells from irrelevant peptide–treated mice (7,290.03 ± 882.79 cpm) (P < 0.01). Taken together, these results show that CD4+CD25+ Treg cells may play a critical role in altered HA308–317 peptide–induced suppression of CIA.

Figure 5.

Enhanced in vitro suppressive function of CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice with CIA. A, CD4+CD25− T cells from mice with CIA were cocultured with antigen-presenting cells (APCs) in the presence or absence of CD4+CD25+ Treg cells from mice treated with wild CII263–272 peptide, wild HA308–317 peptide, altered HA308–317 peptide, or irrelevant peptide and stimulated with 10 μg/ml wild CII263–272 peptide. Proliferation was measured after 4 days of culture. ∗∗ = P < 0.01 versus medium (CD4+CD25− T cells from mice with CIA cocultured with APCs, without CD4+CD25+ Treg cells added). ΔΔ = P < 0.01 versus CD4+CD25+ Treg cells from mice treated with irrelevant peptide. ##= P < 0.01 versus CD4+CD25+ Treg cells from mice treated with wild CII263–272 peptide or wild HA308–317 peptide. B, CD4+CD25− T cells were cocultured with APCs in the presence of CD4+CD25+ Treg cells derived from altered HA308–317 peptide–treated mice. We then added neutralizing antibodies against interleukin-10 (anti–IL-10) or transforming growth factor β (anti-TGFβ) or the respective isotype control antibodies. ∗∗ = P < 0.01 versus no antibody. Results are expressed as the mean ± SD of 3 separate experiments (6 mice per group per experiment). Teff = effective T cells (CD4+CD25− T cells) (see Figure 1 for other definitions).

To evaluate the contribution to the suppression of CIA of IL-10 and TGFβ produced by CD4+CD25+ Treg cells from mice treated with altered HA308–317 peptide, CD4+CD25− T cells from mice with CIA stimulated with wild CII263–272 peptide in vitro were cocultured with APCs in the presence of CD4+CD25+ Treg cells isolated from altered HA308–317 peptide–treated mice with CIA and inhibitory concentrations of anti–IL-10 or anti-TGFβ mAb. Results revealed that the neutralization of either IL-10 or TGFβ could restore proliferation of CD4+CD25− T cells compared with no antibody (mean ± SD 7,245.37 ± 824.42 cpm for anti–IL-10 versus 3,328.91 ± 701.32 cpm for no antibody; P < 0.01) (6,735.63 ± 874.37 cpm for anti-TGFβ versus 3,328.91 ± 701.32 cpm for no antibody; P < 0.01) (Figure 5B). These findings indicate that both IL-10 and TGFβ contribute to the suppressive function of CD4+CD25+ Treg cells.

Greater therapeutic effects on CIA of CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice.

To further explore whether the function of Treg cells was enhanced by altered HA308–317 peptide in vivo, we performed adoptive cell transfer experiments. Mice with CIA received CD4+CD25+ Treg cells isolated from mice with CIA treated with altered HA308–317 peptide, wild HA308–317 peptide, wild CII263–272 peptide, or irrelevant peptide. As shown in Figure 6A, transfer of CD4+CD25+ Treg cells from mice treated with irrelevant peptide resulted in continued severe arthritis in recipient mice with CIA (mean ± SD arthritis score 7.92 ± 1.39 on day 27). CD4+CD25+ Treg cells from wild HA308–317 peptide–treated and wild CII263–272 peptide–treated mice slightly decreased the severity of arthritis (arthritis scores of 5.67 ± 0.82 and 5.99 ± 1.27, respectively; both P < 0.01). However, when transferred cells were from altered HA308–317 peptide–treated mice, the severity of arthritis was significantly reduced (arthritis score 4.08 ± 0.86) compared with that of wild HA308–317 peptide–treated and wild CII263–272 peptide–treated mice (both P < 0.05), indicating that CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice possessed stronger regulatory capacity.

Figure 6.

Therapeutic effects of CD4+CD25+ Treg cells from altered HA308–317 peptide–treated mice on CIA. A, Mice with CIA received CD4+CD25+ Treg cells isolated from mice with CIA treated with irrelevant peptide, wild CII263–272 peptide, wild HA308–317 peptide, or altered HA308–317 peptide. Development of arthritis was monitored for 27 days, and mean ± SD arthritis scores were obtained every 3 days. Analysis of variance with Bonferroni adjustment was used for multiple comparisons. B, Serum was collected from recipient mice and analyzed for concentrations of interleukin-6 (IL-6), tumor necrosis factor α (TNFα), IL-17, IL-10, and transforming growth factor β (TGFβ). Results are expressed as the mean ± SD (10 mice per group). #= P < 0.05; ##= P < 0.01 for whole differences among 4 groups at each time point. ∗ = P < 0.05; ∗∗ = P < 0.01 versus CD4+CD25+ Treg cells from irrelevant peptide–treated mice. Δ = P < 0.05; ΔΔ = P < 0.01 versus CD4+CD25+ Treg cells from wild CII263–272 peptide–treated or wild HA308–317 peptide–treated mice. See Figure 1 for other definitions.

Furthermore, adoptive transfer of CD4+CD25+ Treg cells derived from wild HA308–317 peptide–treated, wild CII263–272 peptide–treated, or altered HA308–317 peptide–treated mice resulted in increased production of the regulatory cytokines IL-10 and TGFβ in recipient mice concomitant with the reduction of IL-17, IL-6, and TNFα, and CD4+CD25+ Treg cells derived from altered HA308–317 peptide–treated mice had stronger effects than those derived from wild HA308–317 peptide–treated and wild CII263–272 peptide–treated mice (Figure 6B). Collectively, the results demonstrated that the altered HA308–317 peptide–mediated suppression of CIA through CD4+CD25+ Treg cells occurs via a mechanism involving IL-10 and TGFβ.

DISCUSSION

The immunotherapeutic effects of altered antigen peptide on autoimmune diseases have been extensively studied in recent years (24), including experimental autoimmune encephalomyelitis (25, 26) and type 1 diabetes (27). Previous studies from our laboratory and others have shown that altered CII peptides prevent the development of CIA by regulating the differentiation of T helper cell subsets (28–32). HA308–317 peptide is a high-affinity HLA–DR4/1–binding peptide. Skinner et al have described that HA peptide induced T cell activation in an HLA–DR4/1–restricted manner and thus might mediate the inflammatory process in RA (33). By computer modeling analysis, our altered HA308–317 peptide has been shown to have an affinity for binding of HLA–DR1 similar to that of wild HA308–317 peptide, which has 10-fold higher affinity to HLA–DR1 than does wild CII263–272 peptide (34). Furthermore, with the substitution of TCR contact residues, altered HA308–317 peptide could inhibit T cell activation induced by wild HA308–317 peptide or wild CII263–272 peptide (18). Thus, altered HA308–317 peptide, which not only can bind to HLA–DR1/4 molecules but also can inhibit responses at the TCR level, might be useful in suppression of autoimmunity.

This is the first study to elucidate the effects of altered HA308–317 peptide therapy in mice with CIA. A number of novel findings emerged from this study, with several major implications. One of the important findings is that wild CII263–272 peptide, wild HA308–317 peptide, and altered HA308–317 peptide could induce potent immune tolerance by intranasal administration, but altered HA308–317 peptide had even stronger inhibitory effects than wild HA308–317 peptide and wild CII263–272 peptide. Recently, antigen-specific mucosal tolerance has been reported in CIA (35, 36). In the present study, we demonstrated that intranasal treatment with wild CII263–272 peptide, wild HA308–317 peptide, or altered HA308–317 peptide ameliorated clinical signs in mice with CIA. Several possible mechanisms may explain the therapeutic effects of these 3 peptides in vivo, including the suppression of effector Th1/Th17 cells and the induction of Treg cells. Th17 cells are thought to play an important role in the pathogenesis of RA and CIA (37). Our data showed that these 3 peptides substantially reduced Th17 cells and IL-17 levels in mice with CIA. In addition, we observed that circulating IL-6 levels were significantly decreased in mice treated with these 3 peptides. The reduction in IL-6 production could prevent Th17 cell development and promote Treg cell generation in the presence of TGFβ, which helped to ameliorate CIA (38).

The induction of adaptive Treg cells may be another mechanism responsible for the tolerogenic effects of these 3 peptides in CIA. Our results showed that the frequency of CD4+FoxP3+ T cells did increase significantly in mice treated with these 3 peptides. Adoptive transfer of CD4+CD25+ T cells from mice with CIA treated intranasally with these 3 peptides substantially reduced arthritis, which implied that the function of Treg cells was enhanced by these peptides. Meanwhile, CD4+CD25+ T cells from mice with CIA that received intranasal administration of altered HA308–317 peptide could suppress effector T cell responses in vitro. Taken together, our data indicated that these 3 peptides exerted their protective effect against CIA at least in part by increasing the frequency of Treg cells and enhancing their function. There are a number of mechanisms by which Treg cells dampen the immune system, including the important involvement of the cytokines TGFβ and IL-10 (39). We demonstrated that both IL-10 and TGFβ contributed to the suppressive function of CD4+CD25+ T cells in CD4+ T cell activation, since anti–IL-10 or anti-TGFβ mAb could significantly attenuate the protective effect of Treg cells derived from altered HA308–317 peptide in vitro.

The greater therapeutic effect of altered HA308–317 peptide on CIA in vivo, compared with that of wild CII263–272 peptide and wild HA308–317 peptide, is most interesting. One possible explanation might be that the low affinity of the TCR–peptide binding resulted in the preferential induction of Treg cells instead of activation of effector T cells (40). The study by Sauer et al delineated signaling events that control the de novo expression of FoxP3 in naive peripheral CD4+ T cells and in thymocytes. Those investigators demonstrated that weak TCR signaling conferred FoxP3 expression and Treg cell–like gene expression profiles. Conversely, continued TCR signaling extinguished the ability of naive CD4+ T cells to induce FoxP3 (40). We speculate that the lower altered HA308–317 peptide–TCR binding affinity that resulted from the substitutions of TCR contact residues may have led to weaker TCR signaling, thus conferring greater production and enhanced function of Treg cells compared with that conferred by wild CII263–272 peptide and wild HA308–317 peptide.

Another important finding of our study was the opposite effects of altered HA308–317 peptide and wild HA308–317 peptide on T cell activation in vitro. We used both 3H-thymidine and CFSE to examine the proliferation of CD4+CD25− T cells. CD4+CD25− T cells exhibited significantly stronger proliferative ability when stimulated by wild HA308–317 peptide than when stimulated by irrelevant peptide. However, compared with irrelevant peptide, altered HA308–317 peptide inhibited CD4+CD25− T cell proliferation. Consistent with the finding in vivo, altered HA308–317 peptide displayed striking immunomodulating effects on the effector memory CD4+ T cells in vitro. It resulted in the skewing of activated CD4+ T cells toward a lower frequency of Th1/Th17 cells but toward an increased proportion and activity of CD4+FoxP3+ T cells as well as increased levels of the antiinflammatory cytokines IL-10 and TGFβ. In contrast, in vitro wild HA308–317 peptide had effects on CD4+ T cells opposite to those of altered HA308–317 peptide, with increased Th1/Th17 cell populations and decreased Treg cell populations, which was concordant with the findings of Skinner et al as well as with those of our previous studies (18, 33). Thus, these findings indicate that altered HA308–317 peptides with substitutions of TCR contact residues are marginal stimulators of T cells and are able to competitively inhibit arthritic T cell activation and selectively promote Treg cell differentiation in vitro.

There is evidence that indoleamine 2,3-dioxygenase (IDO) plays a major role in the generation and function of Treg cells, although multiple factors are involved in this process (41). Our preliminary results indicated that wild HA308–317 peptide–stimulated or wild CII263–272 peptide–stimulated CD4+CD25− T cells obtained significantly stronger proliferative ability when the inhibitor of IDO (1-methyl-D-tryptophan) was applied. In the altered HA308–317 peptide treatment group, the application of 1-methyl-D-tryptophan also enhanced the proliferative ability of CD4+CD25− T cells, although this did not reach statistical significance (data not shown). Further studies are warranted to elucidate the possible mechanisms underlying the increased induction of Treg cells observed in our current study.

In conclusion, therapeutic intranasal treatment with wild CII263–272 peptide, wild HA308–317 peptide, or altered HA308–317 peptide could ameliorate CIA by inducing Treg cells and suppressing the development of CD4+ effector T cells, including Th1 and Th17 cells, but altered HA308–317 peptide exerted more potent therapeutic effects than either wild HA308–317 peptide or wild CII263–272 peptide. Thus, altered HA308–317 peptide could be considered a promising candidate for RA treatment.

AUTHOR CONTRIBUTIONS

All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Zhanguo Li had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study conception and design. J. Sun, R. Li, Guo, Jia, X. Sun, Liu, Y. Li, Lu, Z. Li.

Acquisition of data. J. Sun, R. Li, Jia, X. Sun, Liu, Y. Li, Z. Li.

Analysis and interpretation of data. J. Sun, R. Li, Guo, Jia, X. Sun, Liu, Y. Li, Huang, Lu, Z. Li.

Ancillary