|Home | About | Journals | Submit | Contact Us | Français|
Treatment of (NZB × NZW)F1 (NZB/W) lupus-prone mice with the anti-DNA Ig-based peptide pCons prolongs the survival of treated animals and effectively delays the appearance of autoantibodies and glomerulonephritis. We have previously shown that part of these protective effects associated with the induction of CD4+CD25+Foxp3+ regulatory T cells (Tregs) that suppressed autoantibody responses. Since the effects of pCons appeared secondary to qualitative rather than quantitative changes in Tregs, we investigated the molecular events induced by tolerance in Tregs and found that signaling pathways including ZAP70, p27, STAT1, STAT3, STAT6, SAPK, ERK and JNK were not significantly affected. However, peptide tolerization affected in Tregs the activity of the mitogen-activated protein kinase p38, whose phosphorylation was reduced by tolerance. The pharmacologic inhibition of p38 with the pyridinyl imidazole inhibitor SB203580 in naïve NZB/W mice reproduced in vivo the effects of peptide-induced tolerance and protected mice from lupus-like disease. Transfer experiments confirmed the role of p38 in Tregs on disease activity in the NZB/W mice. These data indicate that the modulation of p38 activity in lupus Tregs can significantly influence the disease activity.
Suppression of effector immune cells by CD4+CD25+Foxp3+ regulatory T cells (Tregs) is a major mechanism of peripheral immune tolerance (1-2). Despite recent progresses in understanding key aspects of the biology of the Tregs, it is largely unknown which molecular mechanisms Tregs employ in their activity (other than upregulation of Foxp3), and what biochemical pathways are modulated in relation to the functional changes that occur in these cells. Indeed, little is known on the molecular pathways that promote or inhibit the activity of Tregs in physiologic and pathologic conditions, despite the many advances in the characterization of Treg phenotypes and suppressive functions (3-4). A better knowledge of these aspects could lead to the development of targeted therapeutic interventions in diseases that are characterized by immune dysregulation and impaired number and/or function of Tregs, such as systemic lupus erythematosus (SLE) (5).
We have previously shown that tolerogenic administration of the anti-DNA peptide pCons induced functional Tregs in NZB/W lupus-prone mice (6). We extend here those findings by showing that phosphorylation of the p38 mitogen-activated protein kinase (MAPK) (p38) is downregulated in Tregs of pCons-tolerized mice.
MAPK's are a group of evolutionarily conserved serine/threonine kinases that are activated in response to a variety of extracellular stimuli and mediate signal transduction from the cell surface to the nucleus (7). Four major types of MAPK cascades have been reported in mammalian cells that respond synergistically to different upstream signals. MAPK's are part of a three-tiered phospho-relay cascade consisting of MAPK, a MAPK kinase (MEK) and a MAPK kinase kinase (MEKK). Controlled regulation of these cascades is involved in cell proliferation and differentiation, and p38 is activated in response to inflammatory cytokines, endotoxins, heat shock and osmotic stress (8).
Our herein described finding of a decreased activation of p38 in tolerized Tregs identifies a pathway modulated by immune tolerance that could be targeted in Tregs in SLE.
Female (NZB × NWZ)F1 (NZB/W) mice were purchased from The Jackson Laboratory (Bar Harbor, ME) or obtained from our colony at UCLA. All animals were treated according to the National Institutes of Health guidelines for the use of experimental animals, with the approval of the UCLA Animal Research Committee for the Use and Care of Animals.
For tolerance induction, 10- to 12-wk-old NZB/W mice received a single i.v. dose of 1 mg of pCons (which contains T cell determinants from different J558 VH regions of NZB/W anti-dsDNA Ig) dissolved in saline (9). Control mice received an identical volume of saline or equal dose of negative control peptide pNeg i.v. (9). There was no significant difference in the percentage and total numbers of Tregs between mice that received saline and pNeg, as reported before (6). Peptides were synthesized at Chiron Biochemicals (San Diego, CA), purified to a single peak by HPLC, and analyzed by mass spectroscopy for expected amino acid content prior to use. One week after treatment, single cell suspensions of splenocytes were prepared by passing cells through a sterile wire mesh. After lysis of RBC with ACK lysing buffer (Sigma-Aldrich, St. Louis, MO), cells were centrifuged, washed, and resuspended in HL-1 medium (BioWhittaker, Walkersville, MD) prior to experimental use.
After cell wash and blockade of Fc-γ receptors, mAb to surface markers or control isotype-matched fluorochrome-labeled Ab in PBS/2% FCS were added for 20 minutes at 4°C. For surface staining, the following fluorochrome-labeled mAb from eBioscience (San Diego, CA) were used: anti-CD3, anti-CD4, anti-CD25, and anti-CD19. For Foxp3 detection, cells were fixed and permeabilized before incubation with anti-Foxp3–PE (eBioscience). Samples were read on a BD FACSCalibur™ and analyzed with FCS Express® (De Novo Software, Thornhill, ON, Canada). For purification of Tregs, sorting was done from splenocytes as CD4+CD25+ T cells by FACSVantage™ (BD Biosciences) or with the Mouse Regulatory T Cell Isolation kit (Miltenyi Biotec, Auburn, CA) using an AutoMACS™ Separator (Miltenyi Biotec). Purity of cells was determined by FACS analysis as >90% Foxp3+ cells among gated CD4+CD25+ T cells. The sorted populations were routinely >95% pure. For signaling proteins, co-staining of cells was done after permeabilization and subsequent use of p38 (C-20) and p-p38 (Tyr182) Abs (Santa Cruz Biotechnology, Santa Cruz, CA) and fluorochrome-conjugated secondary Ab or matched control Ab (eBioscience) with the Cytofix/Cytoperm™ Kit (BD Biosciences), following the manufacturer's instructions.
Western blot analyses were performed as previously described (10). Briefly, total cell lysates from sorted cells were obtained in 50 mM HEPES (pH 7.5), 250 mM NaCl, 1 mM EDTA, 1% Triton X-100, 10 mM sodium fluoride, 1 mM sodium orthovanadate, 2 μg/ml aprotinin, 2 μg/ml leupeptin, and 2 μg/ml pepstatin. Proteins were separated by sodium dodecyl sulfate–polyacrylamide gel electrophoresis (SDS-PAGE) and then transferred onto PVDF membranes (BioRad Laboratories, Hercules, CA). Membranes were blocked in 5% nonfat milk/PBS, 0.5% Tween 20 (PBST) at 4°C for 2 hr, and then incubated with Ab from Cell Signaling Technology (anti-ZAP-70, anti-p27 Kip1, anti-ERK1/2, anti-STAT1, anti-STAT3, anti-STAT6, anti-SAPK, anti-p38) and Santa Cruz Biotechnology (anti-JNK, anti-p38) before being washed in PBST and incubated with peroxidase-conjugated secondary Ab. After an additional wash, peroxidase activity was detected with the ECL system (Amersham, Piscataway, NJ) or Femto system (Thermo Scientific, Rockford, IL). Membranes were stripped and reprobed with anti-phospho Abs from Cell Signaling Technologies (anti-p-ZAP-70Tyr319, anti-p-ERK1/2Thr202/Tyr204, anti-pSTAT1Ser727, anti-STAT3Ser727, anti-STAT6Tyr641, anti-p-SAPKThr183/Tyr185, anti-p-p38Thr180/Tyr182) and Santa Cruz Biotechnology (anti-p-JNKThr 183/Tyr 185, anti-p-p38Tyr182) and again stripped and reprobed with anti-β-actin Ab (Cell Signaling Technology), to determine equivalency of loading. Exposed films were quantified by densitometric band analysis with the ScionImage program (Scion Corporation, Frederick, MD).
ELISA for IL-1 and TNF-α were performed using commercial ELISA kits (R&D Systems, Minneapolis, MN) following the manufacturer's instructions. IgG and anti-dsDNA Ab were analyzed by standard ELISA, as described elsewhere (6, 11).
The p38 inhibitor 4-(4-fluorophenyl)-2-(4-methylsulfinylphenyl)-5-(4-pyridyl)1H-imidazole (SB203580) and negative control molecule 4-ethyl-2(p-methoxyphenyl)-5-(4′-pyridyl)-IH-imidazole (SB202474) were purchased from Calbiochem (San Diego, CA) and dissolved in saline. For in vivo treatment, mice were injected i.p. daily for 2 weeks with 2 mg/kg SB203580 or SB202474 or equal volume of saline (12).
Sorted Tregs, CD4+CD25− T cells, or B cells from treated mice or controls were injected into irradiated recipient mice, as described previously (11). Briefly, each recipient mouse received 600–800 rads two hours prior to an i.v. injection of 10:10:1 ratio of Tregs (1 × 107), CD4+CD25− (effector) T cells (1 × 107), B cells (1 × 106), or isolated sorted lymphocyte subsets or Tregs plus effector T cells suspended in PBS. After transfer, sera of recipient mice were monitored weekly for IgG and anti-dsDNA Ab by ELISA. Early morning urine was monitored at 2 wk intervals for presence of proteinuria using Albustix strips (Bayer, Elkhart, IN), and animal survival was analyzed using Kaplan-Meier curves with Prism 4 software (GraphPad, San Diego, CA).
Analyses for statistical significance were performed using Prism 4 software (GraphPad). Parametric testing between two groups was performed by the paired t test or by the Mann-Whitney U test. Nonparametric testing among more than two groups was performed by one-way ANOVA. Values of P<0.05 were pre-specified as significant.
We previously reported that NZB/W mice tolerized with the anti-DNA Ig peptide pCons were protected from the development and progression of renal disease (9). The tolerized mice had significantly reduced serum titers of anti-DNA Ab and decreased production of pro-inflammatory cytokines when compared to control mice (9). Since part of the protective effects could be explained with the suppression of anti-DNA responses by pCons-induced CD4+CD25+Foxp3+ regulatory T cells (6), we studied the quantitative and qualitative characteristics of these regulatory T cells in relation to tolerization with pCons. The number of Tregs in pCons-tolerized and control NZB/W mice did not change significantly after treatment both as percentage (Figure 1a) and total number of cells (Figure 1b). This finding could not explain the observation of an increased suppressive efficacy of Tregs after pCons-induced immune tolerance (6). Therefore, we asked whether the lack of quantitative effects on Tregs was concomitant to the presence of qualitative change in these cells. Equal numbers of purified Tregs from tolerized mice or control donor mice were adoptively transferred into syngeneic premorbid NZB/W recipients (16 weeks-old) to monitor in vivo the activity of transferred Tregs on development of renal disease. Mice that had been administered Tregs from tolerized mice had delayed development of proteinuria (P<0.04) and increased survival (P<0.001) when compared to mice given Tregs from controls (Figure 2). These data suggest that tolerization of NZB/W mice with pCons can induce qualitative, rather than quantitative changes in Tregs.
Since Tregs from tolerized mice were more effective than Tregs from controls to suppress lupus-like disease (ref. 6, and Figure 2) despite a lack of quantitative changes between the two groups of mice (Figure 1), we investigated whether the differences between the two groups of animals could be explained by a modulation in molecular signalling events in Tregs. Immunoblot analyses indicated no difference in phosphorylation of the ζ-chain-associated tyrosine kinase ZAP70, suggesting that initiation of TCR-proximal T-cell signaling events was not significantly affected by tolerance with pCons (Figure 3). The p27-dependent cell cycle activity in Tregs was also not influenced by tolerance induction, as no changes were found in the expression of the cycline-dependent kinase inhibitor protein p27 (Figure 3). When Signal Transducers and Activators of Transcription (STATs) were studied for the responsiveness to cytokine signalling and/or cell growth, no differences in activation of STAT1, STAT3 and STAT6 were detected in the Tregs after peptide-induced tolerance (Figure 3).
Next, the MAPK/ERK pathway was analyzed because in T cells this pathway couples intracellular responses to binding of growth factor, including cytokines, to cell surface receptors. This pathway also promotes T cell differentiation, proliferation and responsiveness to cytokines and stress stimuli. No significant differences were observed between Tregs from tolerized and control mice in the activation of JNK and the JNK/stress-activated protein kinase (SAPK), and in the activation of extracellular signal-regulated kinases (ERK) (Figure 3). The only significant difference between Tregs from tolerized mice and control mice was a decreased phosphorylation of p38 after treatment with pCons (Figure 3). Interestingly, there was no difference in the phosphorylation of p38 between CD4+CD25− T cells before and after tolerance, indicating that pCons influenced p38 activity in the Tregs but not in CD4+CD25− T cells (Figure 3). These findings were confirmed by showing a reduced activation of p38 in Tregs from tolerized mice in flow cytometry experiments (Figure 4).
To address whether pharmacologically-induced downregulation of p38 (which would mimic pCons-induced downregulation of p38) could affect disease in NZB/W lupus mice, we treated naïve animals with the p38 inhibitor SB203580, the control molecule SB202474, or vehicle, and followed treated animals for survival and for development of anti-DNA Ab. Mice that received SB203580 (that downregulated p38 activity in Tregs in ex vivo flow cytometry experiments, data not shown) had a significantly reduced titer of anti-DNA Ab (P<0.04) when compared to control mice treated with SB202474 or vehicle (Figure 5a). Moreover, inhibition in vivo of p38 also resulted in an increased survival of mice (P<0.003) as compared to controls treated with SB202474 or vehicle and, interestingly, none of the animals treated with SB203580 had died six months after treatment (Figure 5b).
Considering that phosphorylation of p38 results in pro-inflammatory activity, we also addressed whether a reduced activation of p38 could associate with a change in the expression of IL-1 and TNF-α, which are cytokines that contribute to chronic inflammatory processes in systemic autoimmunity (13). Both IL-1 and TNF-α were reduced in sera of mice treated with SB203580, as compared to control mice (Figure 6). These data indicate that the in vivo effect of SB203580 on p38 inhibition associates with a reduced production of inflammatory cytokines and beneficial effects on disease in NZB/W mice.
Both SB203580 and peptide tolerization inhibit p38 activity. We wondered whether cumulative effects would derive from combining pharmacologic and tolerogenic treatments. The reduced phosphorylation of p38 by pCons peptide in Tregs was decreased further by co-treatment with the p38 inhibitor (P<0.01, Figure 7a). The inhibition of p38 in vivo did not affect the number of peripheral Tregs when compared to control treated animals (Figure 7b) or Foxp3 expression (not shown).
Next, we performed in vivo transfer of Tregs with downregulated p38, following a strategy that allows to monitor anti-DNA Ab production in vivo (11). Briefly, mice were irradiated before transfer with Tregs, T helper (Th) cells and B cells. The Tregs were derived from tolerized mice treated with either p38 inhibitor or controls, the Th cells were derived from untreated mice, and the B cells were from old mice. The combination of Th and B cells served as positive controls for in vivo production of anti-DNA Ab (11). As shown in Figure 8, mice receiving Tregs with inhibited p38 displayed reduced IgG and delayed proteinuria than controls.
This study shows that the identification of T-cell signaling events in adaptive Tregs induced by tolerance with peptide can identify targets for pharmacologic modulation of murine SLE. We had previously shown that high doses of the anti-DNA Ig-based peptide pCons effectively suppressed autoimmunity in lupus-prone NZB/W mice and that one mechanism responsible for the protection was the induction of functional Tregs, in addition to inhibitory CD8+ T cells (6, 9, 12, 14-15). However, pCons had little effect on the number of peripheral Tregs (Figure 1). Since an increased suppressive capacity of Tregs after tolerance (Figure 2) had to be explained by qualitative rather than quantitative changes, we investigated whether altered intracellular signaling events in Tregs from pCons-treated mice might associate with changes in Treg activity. In conventional T cells, the intracellular signaling events that follow TCR stimulation result in the activation of tyrosine kinases and the assembly of scaffolds of adaptor molecules whose phosphorylation leads to activation of downstream effectors including serine/threonine kinases such as MAPK and the activation of transcription factors including nuclear factor (NF)-κB (16). The activation of these cascades can engage a variety of T cell functions including the production of cytokines and cell-cycle progression. While most signaling studies have typically focused on conventional T cells, some studies have also investigated the intracellular signaling in Tregs, and the importance of intracellular signaling in the biology of Tregs as well as its relevance to possible therapeutic applications has been recognized (17). As Tregs express a TCR, they require signaling to NF-κB through IKK2 (18) and, although hyporesponsive to antigenic stimulation (3-4), Tregs can signal. Tregs could nonetheless have specific signaling that would make them respond to stimuli differently than conventional activated T cells (19). For example, Tregs are in an anergic state concomitant with an intact proliferative potential, as indicated by the observation that Tregs unable to flux Ca2+ proliferate in response to lymphopenia after TCR engagement (20). Thus, cells remain anergic but maintain an intact suppressive function. The anergic state of Tregs may not represent a default state but rather an actively maintained gene program (19). Signaling requirements of Tregs might reflect the fact that these cells are activated or memory T cells that can react to self antigens, in our case pCons (21-22), and the signaling pathways in the Tregs could contribute to their maintenance and function.
Since cytokine expression can contribute to the development, maintenance and differentiation of most immune cells, it would have been possible that the modulated function of Tregs after tolerance could have associated with differences in the expression of STATs. In this context, STAT1 is involved in upregulating genes related to signaling by either type I or type II interferons (IFNs), and STAT3 is activated in response to various cytokines and growth factors including IFNs. Both type I and II IFNs have been linked to the pathogenesis of SLE in humans and in animal models of the disease, and a type I IFN signature is characteristic of about 50% of SLE patients (23-24). However, our analysis on phosphorylation of STATs did not find significant differences in the expression of these transcription factors, suggesting that these pathways may not affect significantly the activity of Tregs after pCons-induced tolerance in NZB/W mice. Also, the lack of significant changes in p27 after tolerance with pCons indicated that the effects of pCons on Tregs did not influence cellular expansion, in line with the observations reported in Figure 1, and the upstream events to the TCR (ZAP70) were unchanged by tolerance with pCons. On the other hand, events not affected by mitogenic stimuli and related to cellular stress and modulated by MAPK may suggest that decreased activation of p38 in Tregs of pCons-tolerized mice might represent a sensing mechanism to modulate inflammation in mice poised to develop systemic autoimmunity.
Another consideration relates to the possibility that the modulated activity of p38 in Tregs after peptide-induced tolerance may not exclude the contribution of intermediate immune cell(s) and/or soluble factor(s). We showed that tolerance with pCons in NZB/W mice associated with a quantitative expansion of CD8+ T cells (16) that suppressed anti-DNA Ab (11, 14-15), and Sharabi and Mozes recently showed that inhibitory/suppressor CD8+ T cells are required for the optimal function/induction of Tregs after anti-DNA peptide-induced tolerance in mice (25). Other authors found that dendritic cells can influence activity and function of Tregs (26) and convert Foxp3− T cells into Foxp+ T cells (27). Additionally, the requirement of soluble factors such as IL-10 has been described in the generation (28) and/or activity of Tregs (29), and p38 signaling has been associated with TGF-β-mediated conversion of polyclonal CD4+CD25− T cells into Tregs (30). Investigations are needed to clarify the role of these players in this system to define how pCons can influence the quality (function) and/or quantity (conversion) of Tregs from pre-existing cells. At present, our findings identify p38 as a molecule modulated by peptide tolerance in Tregs and help to explain, at a molecular level, some of the beneficial effects caused by p38 inhibition in animal models of arthritis (31-32) and SLE (33) through a novel link between modulation of p38 and Tregs.
To conclude, we have identified through a model of induced tolerance a molecular signaling pathway that allows targeted manipulation of Tregs through pharmacologic intervention. This system could represent a new tool to modulate Treg-mediated suppression in SLE.
This work was supported by the National Institutes of Health grants AR53239 (to A.L.C.), AI065645 and AR054034 (to R.P.S.), AI46776 (to B.H.H.), the Arthritis Foundation Southern California Chapter (to A.L.C.), and the Arthritis National Research Foundation (to E.V.L.). G.M. is supported by the ERC-Starting Grant 202579 and by JDRF-Telethon Grant GJT08004.
The authors have no financial conflict of interest.