IL-28 Supplants Requirement for Treg Cells in Protein σ1-Mediated Protection against Murine Experimental Autoimmune Encephalomyelitis (EAE)

Conventional methods to induce tolerance in humans have met with limited success. Hence, efforts to redirect tolerogen uptake using reovirus adhesin, protein sigma 1 (pσ1), may circumvent these shortcomings based upon the recent finding that when reovirus pσ1 is engineered to deliver chicken ovalbumin (OVA) mucosally, tolerance is obtained, even with a single dose. To test whether single-dose tolerance can be induced to treat EAE, proteolipid protein (PLP130–151) was genetically fused to OVA to pσ1 (PLP:OVA-pσ1) and shown to significantly ameliorate EAE, suppressing proinflammatory cytokines by IL-10+ forkhead box P3 (FoxP3)+ CD25+CD4+ Treg and IL-4+CD25−CD4+ Th2 cells. IL-10R or IL-4 neutralization reversed protection to EAE conferred by PLP:OVA-pσ1, and adoptive transfer of Ag-specific Treg or Th2 cells restored protection against EAE in recipients. Upon assessment of each relative participant, functional inactivation of CD25 impaired PLP:OVA-pσ1's protective capacity, triggering TGF-β-mediated inflammation; however, concomitant inactivation of TGF-β and CD25 reestablished PLP:OVA-pσ1-mediated protection by IL-28-producing FoxP3+CD25−CD4+ T cells. Thus, pσ1-based therapy can resolve EAE independently of or dependently upon CD25 and assigns IL-28 as an alternative therapy for autoimmunity.

Secreted by dendritic cells (DCs) and macrophages, IL-28B (IFNl 3) [11][12][13], a newly described member of IFNl family, is known for its anti-inflammatory activity [14]. Sharing a common signaling pathway with anti-viral type I IFNs [12], IL-28's role in EAE has yet to be evaluated, but it can prime tolerogenic DCs in vitro [12]. When adapted as an adjuvant during DNA vaccination, plasmid-encoded IL-28B reduces T reg cell numbers, but enhances granular CD8 + T cells [15]. In this current study, we demonstrate that protection against EAE, mediated by ps1, is conferred by the expected IL-10-producing T reg cells; however, in the absence of functional T reg cells, protection is mediated by IL-28-producing Th2 cells, demonstrating for the first time that Th2 cells produce IL-28, and endogenous IL-28 can confer protection against EAE.
Acquisition of responsiveness to myelin proteins can develop into the autoimmune disorder, multiple sclerosis (MS) [16]. Current MS therapies fail to restore the unresponsiveness to these self-antigens (Ags). While feeding myelin Ags is effective against EAE [3,[17][18][19], when applied to patients, oral feeding with bovine myelin preparations was deemed unsuccessful [20]. Thus, conventional methods to elicit oral tolerance need to be improved. A number of studies have sought to enhance induction of oral tolerance by adapting liposome delivery [21], including oral adjuvants [22] or coupling to mucosal binding molecules [23,24]. Although most of these strategies significantly ameliorate EAE, multiple doses are required to sustain tolerance, lessening the potency of such methods.
Past studies suggest that sustainable tolerance requires the presence of Peyer's patches for initial Ag sampling subsequent oral tolerogen ingestion [25]. Such evidence implicates the importance of Ag-sampling microfold (M) cells to facilitate Ag uptake from the lumenal surface. To direct tolerogen uptake, we hypothesized that M cell adhesins could be employed to target mucosal inductive tissues, as readily induced with a single dose of OVA fused to reovirus ps1, [5,26]. Mucosal OVA-ps1 induced Ag-specific IL-10 + T reg and IL-4 + Th2 cells capable of suppressing immunity to OVA and ps1, even when co-administered with potent mucosal adjuvants [5,26].
Noting the potency of ps1-elicited tolerance, we queried if it could be adapted to treat autoimmunity by genetically fusing two copies of a portion of proteolipid protein (PLP) containing the encephalitogenic sequence (PLP 139-151 ) to OVA-ps1, termed PLP:OVA-ps1. The described studies showed that PLP 139-151induced EAE is ameliorated with a single nasal dose of PLP:OVA-ps1, stimulating the induction of IL-10-producing T reg cells and IL-4-producing FoxP3 + Th2 cells. Notwithstanding that these induced regulatory T cells were entirely protective subsequent their adoptive transfer, and their effects were neutralized by anti-IL-10 receptor (IL-10R) or anti-IL-4 mAb, additional analyses sought to assess alternative regulatory T cell pathways. Functional inactivation of PLP:OVA-ps1-primed T reg cells rendered mice to an aggressive EAE driven by TGF-b-induced Th17 cells. However, PLP:OVA-ps1 could re-confer protection against EAE upon CD25 and TGF-b co-neutralization in a reversible, IL-28-dependent fashion. Thus, these results show that ps1-based therapeutics can stimulate multiple pathways to induce tolerance and, importantly, can be accomplished independently of T reg cells via IL-4 + or IL-28 + Th2 cells.

Results
Nasal PLP:OVA-ps1 Ameliorates EAE Susceptible female SJL mice nasally dosed with PLP:OVA-ps1, OVA-ps1, or PBS were subjected to conventional PLP 139-151 challenge. PBS-and OVA-ps1-dosed mice developed EAE with average clinical scores .3 at peak disease, followed by relapsingremitting disease and never fully recovered ( Figure 1A). Mice dosed with PLP:OVA-ps1 showed delayed development and reduced duration of clinical disease; the average clinical scores at peak disease were ,1. Unlike PBS-or OVA-ps1-dosed mice, PLP:OVA-ps1-protected mice recovered completely from the acute disease following one relapse. PLP 139-151 -specific delayed type hypersensitivity (DTH) responses ( Figure 1B) 2 wks after EAE induction confirmed significant reduction of PLP 139-151 -specific Figure 1. Nasal administration of PLP:OVA-ps1 protects mice from EAE. A. Mice were dosed with 100 mg of PLP:OVA-ps1, OVA-ps1, or PBS on days 221, 214, and 27, challenged with PLP 139-151 peptide on day 0, and monitored daily for the development of clinical disease. PBS and OVA-ps1-dosed mice developed fully pronounced EAE and never recovered completely. PLP:OVA-ps1-immunization prior to EAE induction significantly delayed and ameliorated clinical onset of EAE and resulted in a complete recovery of all PLP:OVA-ps1-dosed mice. Average of 10-15 mice per group is shown. * P,0.05 for PLP:OVA-ps1 vs. PBS. B. Anti-PLP 139-151 DTH response measured two weeks after EAE induction confirmed that, unlike in PBSor OVA-ps1-dosed mice, PLP 139-151 -specific Th1 response was diminished in PLP:OVA-ps1-dosed mice. Mean + SD of 10-15 mice per group is depicted. * P,0.001 for PLP:OVA-ps1 vs. PBS. C. Mice were dosed with a single 100 mg dose of PLP:OVA-ps1 or PBS on day +6 relative to the day of EAE induction. Unlike PBS, single dose administration of PLP:OVA-ps1 significantly inhibited the occurrence and duration of the clinical EAE. * P,0.05 for PLP:OVA-ps1 vs. PBS. D. Mice dosed with PLP:OVA-ps1 or with PBS 21 days before EAE induction were sacrificed at the peak of the disease (day 14 post-challenge), and histopathology of their spinal cords was determined by staining with luxol fast blue (LFB) and H&E. Mice dosed with PLP:OVA-ps1 showed significant reduction in the central nervous system (CNS) tissue pathology (designated by arrows) compared to PBS-dosed mice. * P,0.001 for PLP:OVA-ps1 vs. PBS. doi:10.1371/journal.pone.0008720.g001 Th1 cells by PLP:OVA-ps1-dosed mice, but not those dosed with OVA-ps1, revealing the importance of Ag-specificity induced by ps1-delivered tolerogens.
SJL mice nasally dosed with PLP:OVA-ps1 prior to EAE induction showed minimal mononuclear cell infiltration with reduced demyelination compared to the PBS-dosed group ( Figure 1D; Table 1). FACS analysis performed on spinal cord cells revealed minimal infiltration of inflammatory cells (MHC II + CD45 high ) into the central nervous system (CNS), and when compared with the naive controls, only a negligible percentage of Mac-3 + macrophages was detectable in the CNS (Table 2). In contrast, PBS-dosed mice showed significant CNS infiltration with CD11b + Gr-1 + neutrophils, CD4 + TCR-b + lymphocytes, and Mac-3 + macrophages. PLP:OVA-ps1-mediated protection against EAE is due in part to suppression of encephalitogenic cell infiltration into the CNS.
Single Nasal Dose of PLP:OVA-ps1 Treats EAE To investigate the impact of single PLP:OVA-ps1 dose subsequent EAE challenge, SJL mice were dosed with PLP:OVA-ps1 or with PBS six days after PLP 139-151 challenge. All PBS-dosed mice displayed expected disease with the peak average clinical score of 4 ( Figure 1C). PLP:OVA-ps1-treated mice showed delayed onset of EAE and ameliorated clinical disease with the average clinical score of 1 at the peak of disease ( Figure 1C), implicating the therapeutic potential of ps1-delivered auto-Ags.

IL-10R Blockade Abolishes PLP:OVA-ps1-Derived T reg Cells' Protective Efficacy
To investigate the relative contribution to protection by these PLP-specific T reg and Th2 cells, naive SJL mice were adoptively transferred with PLP:OVA-ps1-derived T reg or Th2 cells and treated with anti-IL-10R mAb 1 day prior and 5 days after EAE induction. Adoptive transfer of Ag-specific T reg cells nearly abrogated EAE, whereas, Th2 cells partially ameliorated disease ( Figure 3A; Figure S1). Anti-IL-10R mAb treatment had no effect upon PLP:OVA-ps1-derived Th2 cells to prevent EAE; however, IL-10R blockade in recipients given PLP:OVA-ps1-derived T reg cells rendered them susceptible to EAE ( Figure 3A). Such pronounced clinical disease by anti-IL-10R-treated T reg cell recipients showed enhanced proinflammatory responses by .11and 9-fold increases in IFN-c and IL-17, respectively, by LN CD4 + T cells when compared to IgG-treated T reg cell recipients ( Figure 3B). Likewise, IL-6 and IL-21 were augmented 3-and 5-fold, respectively, when compared to lymphocytes from T reg cell recipients treated with IgG. Consequently, T reg cells + anti-IL-10R mAb-treated recipients did not produce IL-4, IL-10, or IL-28, but did produce ,2-fold more TGF-b when    compared to T reg cell recipients treated with IgG. Lymphocytes from IgG-treated T reg cell recipients resembled PLP:OVA-ps1treated mice, producing anti-inflammatory cytokines, IL-4, IL-10, and IL-28, and suppressing TGF-b production. Diseased mice from groups treated with PBS + IgG or PBS + anti-IL-10R mAb showed pronounced inflammatory responses evident by augmented IL-6, IL-21, and IL-23, in addition to .5-fold increases in IL-17 when compared to IgG-treated T reg cell recipients ( Figure 3B). IgG-treated Th2 cell recipients produced significantly more IFN-c, at least 3-fold more IL-21 and TGF-b, considerably less IL-10 and IL-28, but similar amounts of IL-4 when compared to IgG-treated T reg cell recipients. Lymphocytes from anti-IL-10R-treated Th2 cell recipients showed inhibition of IL-4, IL-28, and 3-fold less TGF-b when compared to IgG-treated Th2 cell recipients, and yet both groups showed reduced EAE because of their ability to inhibit IFN-c, IL-6, and IL-17.

IL-4 Neutralization Partially Reverses PLP:OVA-ps1-Mediated Protection
Testing the relevance of IL-4 in PLP:OVA-ps1-mediated protection against EAE, groups of mice were dosed with PLP:OVA-ps1 or with PBS on days -14 and -7, subsequently treating them with an anti-IL-4 mAb or rat IgG on days 21 and +5 relative to EAE challenge. Mice dosed with PBS+IgG developed typical EAE onset ( Figure 3C). IL-4 neutralization accelerated onset of clinical disease and amplified disease severity in PBS-dosed mice ( Figure 3C; Table S2). PLP:OVA-ps1 + anti-IL-4 mAb-treated mice also showed earlier disease onset and greater EAE severity than in PLP:OVA-ps1 + IgGdosed mice, but were less pronounced than in PBS + IgG-dosed control mice.
CD4 + T cells isolated from anti-IL-4 mAb-treated and PLP:OVA-ps1-dosed mice at the peak of the disease showed pronounced proinflammatory cytokines, IFN-c, IL-6, and IL-17, Figure 3. PLP:OVA-ps1 protects against EAE via IL-10-producing T reg and IL-4-producing Th2 cells. Mice were dosed with PLP:OVA-ps1, and two weeks later, T reg and CD25 2 CD4 + Th2 cells were isolated and adoptively transferred to naive recipients. On day 0, mice were induced with EAE and on days 21 and +5 received anti-IL-10R mAb or IgG isotype control Ab. A. In contrast to T reg cells + IgG-treated mice, mice given T reg cells + anti-IL-10R mAb developed clinical EAE. Administration of anti-IL-10R mAb had no effect on the clinical disease in mice adoptively transferred with PLP:OVA-ps1-derived Th2 cells. * P,0.05 vs. PBS + IgG. B. Mice were sacrificed at the peak of clinical disease, and CD4 + T cells isolated from their LNs were evaluated for production of cytokines by ELISA. T reg cells + anti-IL-10R-treated mice produced significantly more proinflammatory cytokines and less anti-inflammatory cytokines when compared to T reg cells + IgG-treated mice. Mean and SD of 5 mice per group is depicted; * P,0.05 vs. T reg cells + IgG. C. Mice dosed with PLP:OVA-ps1 or PBS on days 214 and 27 were injected with anti-IL-4 mAb or rat IgG on days 21 and +5. PBS + anti-IL-4-dosed mice developed accelerated and more severe EAE than PBS + IgG-dosed mice. The disease in PLP:OVA-ps1 + anti-IL-4-dosed mice was less severe than in PBS + IgG-dosed mice, but significantly more severe than in PLP:OVA-ps1 + IgG-dosed mice. * P,0.05 for PLP:OVA-ps1 + IgG vs. PBS + IgG or PLP:OVA-ps1 + anti-IL-4. doi:10.1371/journal.pone.0008720.g003 and reduced IL-10 when compared to PLP:OVA-ps1 + IgG-dosed mice ( Figure S2). Overall, IL-4 neutralization in either PBS-or PLP:OVA-ps1-dosed mice induced more IFN-c, when compared to their respective IgG-treated control mice, and significantly less IL-10 was produced in anti-IL-4 mAb-treated PLP:OVA-ps1dosed mice than in PLP:OVA-ps1 + IgG-protected mice.
Functional Inactivation of CD25 + T Cells Abrogates PLP:OVA-ps1-Mediated Tolerance and Stimulates TGF-b-Dependent EAE Adoptive transfer of T reg cells from PLP:OVA-ps1-primed mice was completely protective against EAE challenge ( Figures 3A and S1), further implicating relevance of T reg cells for protection. Neutralization of functional T reg cells did not affect course of EAE in PBS-dosed mice when compared to PBS + IgG-dosed mice ( Figure 4A); however, EAE accelerated in anti-CD25 mAb-treated PLP:OVA-ps1-dosed mice, resulting in a notably more severe EAE ( Figure 4A). The protective regulatory and Th2-type responses normally induced by PLP:OVA-ps1 were abated subsequent CD25 neutralization. The impact was evident by 3-fold reductions in IL-10, loss of IL-4, and augmented proinflammatory responses noted by 5fold for IFN-c, 8-fold for IL-17, and .10-fold enhancements for IL-6 and IL-21, as well as increased TGF-b ( Figure 4B). PBS-dosed mice treated with anti-CD25 mAb or with IgG developed classic proinflammatory responses, and LN CD4 + T cells isolated from these mice showed elevations in IFN-c, IL-6, IL-17, and IL-21 and considerably less IL-10 and IL-4 when compared to PLP:OVA-ps1 + IgG-dosed mice ( Figure 4B). PLP:OVA-ps1 + IgG-treated mice produced the expected elevations in IL-4 and IL-10 and near neutralization of proinflammatory cytokines in a PLP 139-151 -specific T reg celldependent fashion. Moreover, the lack of protection in CD25neutralized PLP:OVA-ps1-dosed mice was associated with enhanced proinflammatory cytokines and a striking induction of TGF-b. TGF-b and CD25 Co-Neutralization Restores ps1-Mediated Protection against EAE Independent of T reg Cells T reg cell neutralization exacerbated EAE, negating the protective capacity of PLP:OVA-ps1, resulting in enhanced TGF-b production ( Figure 4B) and implicating a proinflammatory role of this cytokine in EAE. To address TGF-b's participation in EAE development subsequent CD25 neutralization, PLP:OVA-ps1and PBS-dosed mice were treated in vivo with anti-TGF-b mAb, anti-CD25 mAb, both, or IgG ( Figure 5A). No difference in onset or disease severity was observed between mice dosed with PBS and treated with the different combinations of mAbs, except those mice dosed with PBS + anti-TGF-b mAb recovered sooner from the acute disease ( Figure 5B). In contrast to PLP:OVA-ps1 + anti-CD25 mAb-treated mice, which developed a very aggressive disease, mice given PLP:OVA-ps1 + any of the remaining Ab treatments developed only very mild disease with the average peak clinical score of ,1.5; all of these mice recovered from acute EAE ( Figure 5C). Thus, co-neutralization of CD25 and TGF-b in PLP:OVA-ps1-treated mice resembled ameliorated disease, as seen in PLP:OVA-ps1 + IgG-dosed mice. These results showed that mice functionally neutralized of their CD25 + T reg cells in the presence of tolerogen develop a more aggressive, TGF-bdependent EAE. The suppressive activity of PLP:OVA-ps1 could only be restored upon co-neutralization of TGF-b. These findings corroborate the results in Figure 3B in which minimal to no TGFb was detected in PBS-treated mice, suggesting that TGF-b has a minimal role in PLP-mediated EAE.

Restoration of Tolerance by PLP:OVA-ps1 upon Co-Neutralization of CD25 and TGF-b Is IL-28-Dependent
Examination of cytokine profiles conducted 10 days post-EAE challenge revealed that PLP 139-151 -restimulated mononuclear cells from PBS-dosed groups produced elevated amounts of IL-6, IL-17, IL-21, and IL-23, and very little to no anti-inflammatory IL-4 and IL-10 ( Figure S3). When compared to PLP:OVA-ps1 + IgGtreated mice, PLP:OVA-ps1 + anti-CD25 mAb + anti-TGF-b mAb-treated mice showed a slight reduction in IL-4, but a marked 7.3-fold reduction in IL-10, which was consistent with the lack of functional IL-10-producing T reg cells. In addition, this treatment also reduced IL-22 by 7.7-fold and showed inhibition of IL-6, IL-17, IL-21, and IL-23 ( Figure S3). IL-28 production was preserved by this treatment; however, when compared to diseased PLP:OVA-ps1 + anti-CD25 mAb-treated mice, IL-28 was enhanced 16.4-fold, but with no change in IL-10, implicating that in the absence of functional T reg cells, PLP:OVA-ps1induced protection can occur via Th2-type cells. IL-28 was not produced in diseased mice treated with either PBS plus the various mAbs, or with PLP:OVA-ps1 + anti-CD25 mAb, showing the anti-inflammatory property of IL-28 in EAE. Lack of differences in IL-4-secretion between mice dosed with PLP:OVA-ps1 + IgG or PLP:OVA-ps1 + anti-TGF-b + anti-CD25 mAb further suggests that the presence of T reg cells is not required for IL-4 production ( Figure 5D; Figure S3).
Although T reg cells were absent in mice treated with anti-CD25 plus anti-TGF-b mAbs, upon tolerance induction with PLP:OVA-ps1, alternative regulatory T cells were induced evident by the expression of FoxP3 by CD25 2 CD4 + T cells (Table 3). In fact, the inclusion of anti-TGF-b mAb in the treatment paradigm restored elevated FoxP3 expression by CD25 2 CD4 + T cells to levels similar to those obtained from PLP:OVA-ps1 + IgG-dosed mice. In contrast, PLP:OVA-ps1 + anti-CD25 mAb-treated mice displayed a 67% reduction in FoxP3 + CD25 2 CD4 + T cells when compared to PLP:OVA-ps1-protected mice (Table 3). PLP:OVA-ps1 + anti-TGF-b mAb treatment showed only a 20% reduction in FoxP3 + CD25 2 CD4 + T cells; these mice still retained their T reg cells.
Aside from TGF-b, IL-23 also has been shown to activate Th17 cells [27][28][29]. To determine the necessity of IL-23 for induction of EAE in PLP:OVA-ps1 + anti-CD25 mAb-treated mice, mice were co-treated with anti-IL-23p19 serum. Neutralization of IL-23 had no effect on clinical onset and severity of EAE ( Figure 6).

Endogenous IL-28 Protects against EAE in the Absence of T reg Cells
To investigate the role of IL-28 in the PLP:OVA-ps1-mediated protection against EAE in the absence of functional T reg cells, mice dosed with PLP:OVA-ps1 or PBS were treated with anti-IL-28 rabbit serum (RS), normal RS (NRS), and/or with mAbs to CD25 and TGF-b ( Figure 7A). Treatment of PBS-or PLP:OVA-ps1-dosed mice with anti-IL-28 RS did not significantly alter EAE when compared to their respective control groups dosed with NRS ( Figure 7B). Neutralization of IL-28 in concert with co-neutralized CD25 and TGF-b rendered PLP:OVA-ps1-dosed mice susceptible to EAE development, presumably due to the significant reduction in FoxP3 + Th2 cells ( Figure 7C), reaffirming the importance of IL-28-mediated protection in EAE.
As before, CD4 + T cells from PLP:OVA-ps1 + NRS-and PLP:OVA-ps1 + anti-CD25 + anti-TGF-b-treated mice produced predominantly anti-inflammatory cytokines ( Figure 7D). Although PLP:OVA-ps1 + anti-IL-28-dosed mice conferred protection against EAE ( Figure 7B), upon peptide restimulation, their CD4 + T cells produced proinflammatory cytokines, IFN-c, IL-6, and IL-17 ( Figure 7D), and these mice expressed fewer FoxP3 + Th2 cells ( Figure 7C), unlike PLP:OVA-ps1 + NRS-treated mice that were protected against EAE, but without inducing proinflammatory cytokines. Such differences in protection were not attributed to IL-10 since these remained unchanged in both groups. Neutralization of IL-28 in PBS or PBS + anti-CD25 + anti-TGF-b-dosed mice did not significantly affect proinflammatory cytokine production by CD4 + T cells; however, secretion of IL-4 and TGF-b was significantly reduced in these mice when compared to PBS + NRS-dosed mice. Independent of treatment, all PBS-dosed mice showed only marginal production of anti-inflammatory cytokines. These studies demonstrate IL-28 can be induced to supplant protective IL-10 + T reg cells. To our knowledge, this is the first such report describing IL-28's protective capacity against inflammatory autoimmune diseases.

Discussion
Tolerance is the active inability to respond to self or a defined Ag and represents a method to abolish self-reactivity to ultimately enable protection against autoimmune disease. A caveat in preventing or treating autoimmune diseases in humans is being able to successfully adapt tolerance methods used in various animal models. In a double-blind phase III clinical trial to test the feasibility of inducing oral tolerance to myelin basic protein and PLP, high doses of myelin Ags were administered, and although TGF-bsecreting CD4 + T cells were induced [30], no differences in MS outcomes between placebo and treated were observed [31]. Notwithstanding, alternative methods of mucosal delivery, particularly sublingual, have proven effective in humans to induce tolerance to alleviate allergies to dust mites and grass pollens [32][33][34]. Rendering tolerance via sublingual delivery of these allergens requires contact with the oral mucosa since immediate swallowing of allergens diminishes the tolerogenic capacity [33]. To a limited extent, targeting auto-Ags to the mucosal epithelium has been tested using cholera toxin subunit B (CT-B) to improve diabetes [35] and EAE [36] in mice. However, CT-B also behaves as a mucosal adjuvant [37,38], resulting in conflicting outcomes when Ags chemically coupled to CT-B produce potent Ag-specific immunity [36,39] or induce tolerance [35,40]. Since chemical modification can render a tolerogen immunogenic, the alternative approach of genetically fusing PLP 139-151 to CT-B maintains its ability to remain tolerogenic and suppress EAE [24]. Despite its success, this approach again requires multiple nasal administrations to maintain efficacy against EAE. These collective studies implicate the potential of induced tolerance if given by a suitable mucosal route, allowing for tolerogen uptake. Thus, there have been limited attempts to test the potential of microbial adhesins in humans to facilitate tolerance induction, let alone, with ps1.
To enable this latter possibility and to take advantage of ps1's adhesive properties [5,41], OVA-ps1 was modified with two copies of immunodominant T cell epitope PLP 130-151 , resulting in a functional PLP:OVA-ps1, which, when given nasally, protected susceptible SJL mice against EAE. PLP:OVA-ps1 protection resulted in the diminution of proinflammatory CD4 + T cells and the stimulation of CD4 + T cells producing regulatory and antiinflammatory cytokines. The related protein lacking PLP peptide, OVA-ps1, was not protective, further showing the tolerogenic responses being Ag-specific. Importantly, the therapeutic effect conferred by PLP:OVA-ps1 could be rendered with as low as a single 100 mg dose, similar to that shown for OVA-specific tolerance [5]. While tolerance to encephalitogenic proteins or peptides when applied mucosally can be shown, these generally require multiple and/or large Ag doses to confer protection against EAE [3,17,19,20,42].
Although PLP:OVA-ps1 is a derivative of OVA-ps1, this was done to facilitate monitoring of B cell responses subsequent mucosal delivery of this tolerogen since the PLP 139-151 T cell epitope produced weak to no Ab responses in PBS-treated, PLP 139-151 -challenged mice (data not shown). The intent of this work is to establish the potential of ps1-based therapeutics for MS, and subsequent work will fashion a ps1-based fusion tolerogen bearing relevant human T and B cell epitopes for MS.
Surprisingly, IL-4 deficiency has not been associated with increased susceptibility to autoimmune diseases, suggesting a supportive role for IL-4 in suppression of inflammatory responses against self-Ags [48]. Therefore, consistent with the supportive role of IL-4 in induction of T reg cells [47], possibly IL-4 neutralization downregulates IL-10-producing T reg cells in PLP:OVA-ps1-dosed mice. Inquiry into the role of Th2 cytokines to ps1-dependent tolerance found that the functional inactivation of T reg cells greatly impaired the PLP:OVA-ps1-induced protection against EAE. Anti-CD25 mAb treatment relinquished the tolerogenic property of PLP:OVA-ps1, making it more immunostimulatory, as evident by the increased disease severity and enhanced production of TGFb, IL-6, IL-17, IL-21, and IFN-c with concomitant reductions in IL-4, IL-10, and IL-28. Although IL-13 was enhanced, our preliminary studies implicated lack of proinflammatory function for this cytokine in T reg cell-depleted PLP:OVA-ps1-dosed mice. In contrast to concomitant inhibition of CD25 and IL-23, simultaneous neutralization of TGF-b and CD25 reestablished PLP:OVA-ps1-induced protection, suggesting that anti-CD25 mAb-induced Th17 inflammatory responses in PLP:OVA-ps1dosed mice are mediated via TGF-b rather than IL-23. In addition to its established proinflammatory role [28,29], TGF-b can be a major regulatory cytokine secreted by T reg cells, and its production has been linked to potent suppression of EAE [3,7].
Protected PLP:OVA-ps1 + anti-TGF-b-treated mice showed a modest, although significant, decrease in FoxP3 + Th2 cells when compared to PLP:OVA-ps1 + IgG-dosed mice, but these cells remained significantly enhanced when compared to PLP:OVA-ps1 + anti-CD25-treated diseased mice. Although TGF-b can support conversion of CD25 2 CD4 + T cells to T reg cells via enhanced FoxP3 expression [9,49], our data showed that CD25 neutralization upon PLP:OVA-ps1 treatment results in enhanced TGF-b responses reminiscent of TGF-b's inflammatory properties, perhaps because of the increased presence of IL-6 [29] and/or IL-21 [50].
Mice co-neutralized of the T reg cells and TGF-b, but dosed with PLP:OVA-ps1, reestablished tolerance evidenced by the inhibition of Th1 and Th17 cells and the enhanced production of IL-4, IL-13, and IL-28. Reversion of IL-10 production was not evident presumably because of neutralization of T reg cells. Concomitant neutralization of TGF-b, CD25, and IL-28 reversed the Th2 celldependent tolerance by PLP:OVA-ps1 treatment, resulting in EAE and reaffirming the novel role for IL-28 in protection against EAE in the absence of T reg cells. Combined with their enhanced expression of FoxP3, these results suggest an alternative regulatory pathway that can be induced by PLP:OVA-ps1, but independent of conventional T reg cells. In some cases, minimal levels of IL-10 were still being secreted by CD4 + T cells in PLP:OVA-ps1 + anti-CD25 + anti-TGF-b-dosed mice. Consequently, the possibility of potential synergistic or priming effect by IL-10 upon IL-28 cannot be excluded to account for the observed protection against EAE in the absence of functional T reg cells and proinflammatory TGF-b. A growing body of evidence suggests that type I and II IFNs can induce proinflammatory potential of IL-10 by switching the balance of IL-10 STAT activation from Stat3 to Stat1 [51,52]. Type III IFNs share functional and structural similarities with type I IFNs, including the Jak-STAT signaling pathway [11,13,14,53]; therefore, it is plausible that in the presence of IL-28 the nominal amounts of IL-10 produced by the CD4 + T cells in PLP:OVA-ps1 + anti-CD25 + anti-TGF-b-dosed mice are in fact proinflammatory. The role for IL- 28 has not yet been evaluated in EAE, although IL-28 is known for its anti-inflammatory activity [14], and it has been shown to prime tolerogenic DCs in vitro [12].
In summary, we showed that even a single 100 mg dose of ps1based nasal vaccine in an Ag-specific fashion protected mice against EAE. The ps1-based vaccine protected against EAE via various mechanisms, including activation of IL-10-producing FoxP3 + T reg cells and IL-4-secreting FoxP3 + Th2 cells. In the absence of T reg cells, ps1-based protection against EAE was associated with an increased expression of FoxP3 on CD25 2 CD4 + T cells producing IL-28, which, to our knowledge, is the first report describing regulatory role of IL-28-producing CD4 + T cells conferring protection against EAE. These results show that a single low dose nasal tolerance mediated by genetically modified ps1 can be successfully applied to prevent and/or treat autoimmune diseases.

Ethics Statement
All animal care and procedures were in accordance with institutional policies for animal health and well-being, and approved by MSU Institutional Animal Care and Use Committee.

Preparation of PLP:OVA-ps1
PLP:OVA-ps1 was constructed using the OVA-ps1 backbone [5]. Two copies of PLP peptide (PLP 130-151 ; QAHSLERVC HCLGKWLGHPDKF) separated by the flexible linker (RHRHVDCSGRNLTTLPPGLQE) were synthesized as a single cDNA fragment containing restriction enzyme sites 59 and 39 termini (GenScript Corp.). The synthetic cDNA fragment was amplified by PCR and cloned into pUC19. The 59 and 39 primers encoded EcoRI sites, and 59 primer encoded an ATG initiation codon embedded into an optimal Kozak's sequence. PCR amplified PLP peptides were ligated with the 59 terminus of OVA-ps1 in a pPICZ B vector (Invitrogen Corp.) bearing a histag carboxy terminus for protein purification (Invitrogen), referred to as PLP:OVA-ps1. The junction between the PLP 139-151 epitopes and the OVA-ps1 featured a flexible linker (Gly-Arg-Pro) to minimize steric hindrance between the components. The resulting construct was sequenced and expressed in the yeast Pichia pastoris, according to the manufacturer's directions (Invitrogen Corp.). Recombinant proteins were extracted from yeast cells by a bead-beater (Biospec Products) and purified on a Talon metal affinity resin (BD Biosciences, Palo Alto, CA), according to manufacturer's instructions. Proteins were assessed for purity and quality by Coomassie-stained polyacrylamide gels and by Western blot analysis using a polyclonal rabbit anti-ps1 (produced inhouse) or a polyclonal rabbit anti-OVA Ab (Sigma-Aldrich). All recombinant proteins migrated as a single band with the expected MW.

Mice
Female six wk old SJL mice were obtained from Frederick Cancer Research Facility, National Cancer Institute, and The Jackson Laboratories. All mice were maintained at Montana State University Animal Resources Center under pathogen-free conditions in individual ventilated cages under HEPA-filtered barrier conditions and were fed sterile food and water ad libitum. The mice were free of bacterial and viral pathogens, as determined by antibody screening and histopathological analysis of major organs and tissues.
Tolerance Induction, PLP:OVA-ps1 Treatment, and EAE Challenge For tolerance induction, mice (5-10 mice/group) were nasally dosed up to three times with 50-100 mg of PLP:OVA-ps1 or OVA-ps1 before or 6 days after EAE challenge, as described in the text. Control groups were treated with PBS or equivalent amounts of OVA-ps1. PLP:OVA-ps1 or OVA-ps1 was administered nasally, as previously described [5].

Measurement of Delayed-Type Hypersensitivity (DTH) Responses
To measure OVA-or PLP 139-151 -specific DTH responses [55], OVA or PLP 139-151 (10 mg) was injected into the left ear pinna, and PBS alone (20 ml) was administered to the right ear pinna as a control. Ear swelling was measured 24 h later with an electronic digital caliper (World Precision Instruments). The DTH response was calculated as the increase in ear swelling after antigen injection following subtraction of swelling in the control site injected with PBS.

Histological Evaluation of Spinal Cords
For histological evaluation of tissue pathology, spinal cords were removed 14 days after challenge and fixed with neutral buffered formalin (VWR International), embedded into paraffin, and sectioned at 5 mm. Cross sections of spinal cords were stained with H&E for pathological changes and inflammatory cell infiltration, and adjacent sections with luxol fast blue (LFB) for loss of myelin. Pathological manifestations were scored separately for cell infiltrates and demyelination. Each H&E section was scored from 0 to 4: 0, normal; 1, cell infiltrate into the meninges; 2, one to four small focal perivascular infiltrates; 3, five or more small focal perivascular infiltrates and/or one or more large infiltrates invading the parenchyma; 4, extensive cell infiltrates involving 20% or more of the white matter [7,54]. In each LFB stained section, myelin was also scored from 0 to 4: 0, normal; 1, one small focal area of demyelination; 2, two or three small focal areas of demyelination; 3, one to two large areas of demyelination; 4, extensive demyelination involving 20% or more of white matter [7,54].

FACS Analysis
Lymphocytes from the HNLNs, MLNs, and spleens were isolated 14 days after challenge, and single cell suspensions were prepared, as described above [7]. To obtain lymphocytes from spinal cords, mice were perfused through the left ventricle with 20 ml of ice cold sPBS, and spinal cords were removed by flushing the vertebral canal with media and prepared, as previously described [7].

In Vivo Neutralization of IL-4 and Blockade of IL-10R
To inhibit IL-4 in vivo, mice dosed with PLP:OVA-ps1 on days 214 and 27 before EAE challenge were given i.p. 1.0 mg of anti-IL-4 mAb (clone 11B11; ATCC) on day 21 before challenge, and on day +5 after EAE challenge with PLP 139-151 [56]. Control mice received i.p. injection of 1.0 mg purified rat IgG Ab (AbD Serotec).
To inhibit IL-10 receptor function, mice were i.p. injected with 0.5 mg of anti-IL-10R mAb (clone1B1.3A, BioXCell), or IgG isotype control Ab at the day of an adoptive transfer and 6 days later (day 21 and +5 relative to the day of EAE induction). Mice were induced with EAE on day 0, as described above.

In Vitro T Cell Assays
To assess cytokine production by T reg cells and effector T cells, CD25 + CD4 + and CD25 2 CD4 + T cells (2610 5 ) were stimulated in vitro with anti-CD3 mAb-coated wells (10 mg/ml; BD Pharmingen) and a soluble anti-CD28 mAb (5 mg/ml; BD Pharmingen) for 5 days (final volume of 300 ml in a 48-well plate). Capture ELISA was used to quantify triplicate sets of samples to measure cytokine production [7].

In Vivo Inactivation of CD25 and Neutralization of TGF-b and/or IL-28
Mice were nasally dosed with PLP:OVA-ps1 or PBS on days 214 and 27 relative to the EAE induction with PLP 139-151 . To functionally inactivate CD25 + CD4 + T cells, the same mice were given i.p. 0.5 mg anti-CD25 mAb (clone PC 61.5.3; ATCC TIB-222) on days 25 and 22 before EAE induction. As a control, groups of PLP:OVA-ps1-or PBS-dosed mice received 0.5 mg of purified rat IgG on the same days before EAE challenge. All mice were monitored daily for development of EAE.
To neutralize TGF-b in vivo, mice dosed with PLP:OVA-ps1 or PBS and treated with anti-CD25 mAb or rat IgG were i.p. injected with an additional 0.5 mg of anti-TGF-b mAb (clone 1D11.16.8, ATCC) on days 21 and +5 relative to the day of EAE induction.
To neutralize IL-28, mice were i.p. injected on days 23 and +4 relative to the EAE induction with 1 ml of anti-IL-28 RS (developed in-house by immunizing rabbits with recombinant IL-28B; R&D Systems) or with NRS as control.

In Vivo Neutralization of IL-23
To block IL-23 in vivo, mice dosed with 100 mg of PLP:OVA-ps1 on day 214 and 27 before EAE challenge were given i.p. 0.5 ml of anti-IL-23p19 rabbit serum (RS, made in-house) on day 21 before challenge, and 0.25 ml of anti-IL-23 RS on days 1 and 5 after EAE challenge with PLP 139-151 . Control mice received i.p. injection of an equal amount of NRS (Jackson ImmunoResearch Laboratories).

Statistical Analysis
The ANOVA followed by posthoc Tukey test was applied to show differences in clinical scores in treated vs. PBS mice. The student t test was used to evaluate the differences between variations in cytokine level production, and P-values ,0.05 are indicated, unless specified otherwise.