Regulatory T Cells Attenuate Th17 Cell-Mediated Nigrostriatal Dopaminergic Neurodegeneration in a Model of Parkinson’s Disease

An inciting event that underlies the pathogenesis of Parkinson’s disease (PD) is the accumulation of aggregated proteins within neuronal cell bodies and microglia, which is associated with microglial activation and neuronal death. Deposition of misfolded and nitrated α-synuclein (N–α-syn) into Lewy bodies within nigral dopaminergic neurons of the substantia nigra (SN) pars compacta (1, 2), with subsequent release into extracellular spaces and draining cervical lymph nodes, affects neuronal loss by engaging innate and adaptive immune responses (3–7). This leads to oxidative stress, microglial and APC activation, and neuronal degeneration (3, 6, 7). Interestingly, α-synuclein (α-syn) immunization was shown to generate humoral responses for clearing protein aggregates (8). However, using nitrated forms of α-syn (N-4YSyn) as an immunogen induced profound effector T cell (Teff) responses shown to exacerbate neuroinflammation and neurodegeneration (3), analogous to the untoward T cell-mediated meningoencephalitic responses observed after Aβ immunization (9). Other investigators reported that T cell responses elicited during the course of 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP) intoxication led to accelerated neurodegeneration (5). The molecular and antigenic bases for these T cell-mediated neurodegenerative activities are unclear. However, our previous results demonstrated that nitrated forms of α-syn are recognized by the adaptive immune arm to yield Ag-specific Teffs, which exacerbate neuroinflammation and accelerate nigrostriatal degeneration (3).

In contrast, regulatory components of adaptive immunity affect neural repair and protection. Regulatory T cells (Tregs) protect against MPTP-induced dopaminergic degeneration (10). This raised the specter of opposing effects for CD4⁺ T cell subsets on brain disease; autoaggressive Teff responses speed the tempo of disease, whereas Tregs attenuate neurodegeneration (4, 10–12). This is in keeping with known anti-inflammatory and neurotrophic functions of Treg, their essential role in control of immune-mediated inflammation and regulation of the mononuclear phagocyte phenotype (10, 12–15). Thus, we investigated whether an adjuvant to promote specific regulatory adaptive immune responses could be used with N–α-syn as a vaccine for PD. Using vasoactive intestinal peptide (VIP), a neuropeptide known to induce Treg responses (16, 17), we showed that functional Tregs within N–α-syn splenocytes (SPCs) elicited neuroprotective responses.

Recombinant C-terminal tail of α-syn (4YSyn) was purified, nitrated (N-4YSyn), and tested for endotoxin, as described (3). Male C57BL/6J mice and forkhead box P3 (FoxP3)/GFP C57BL/6J mice (5 wk old, The Jackson Laboratory, Bar Harbor, ME) were primed s.c. with N-4YSyn emulsified in CFA (Sigma-Aldrich, St. Louis, MO), boosted s.c. after 2 wk with N-4YSyn in IFA (Sigma-Aldrich), and cells were harvested 5 d after boost, as described (3). Each milliliter of CFA consisted of 1 mg dry weight heat-killed Mycobacterium tuberculosis (H37Ra, ATCC 25177), 0.85 ml paraffin oil, and 0.15 ml mannide monooleate. IFA contained no mycobacteria. Donor mice that did not receive immunizations were injected i.p. with 15 mg VIP (Sigma-Aldrich) in PBS. Recipient mice received four i.p. injections of vehicle (PBS, 10 ml/kg body weight) or MPTP-HCl (16 mg MPTP, free base in PBS/kg body weight in PBS; Sigma-Aldrich); each injection was given at 2-h intervals. Twelve hours after the last injection, SPCs or Tregs were adoptively transferred to MPTP-intoxicated recipient mice (n = 5–7 mice per group per time point). On days 2 and 7 post-MPTP, mice were sacrificed, and brains were processed for analysis. All animal procedures were in accordance with National Institutes of Health guidelines and were approved by the Institutional Animal Care and Use Committee of the University of Nebraska Medical Center. MPTP safety measures were in accordance with published guidelines (18).

Five days following boost, mice were sacrificed, and single-cell suspensions were prepared from inguinal lymph nodes and spleens. CD4⁺ T cell populations from spleens and lymph nodes were enriched by negative selection with CD4-enrichment columns (R&D Systems, Minneapolis, MN), followed by CD25-PE positive selection with AutoMACS (Miltenyi Biotec, Auburn, CA). As determined by flow cytometric analysis, populations of Tregs and Teffs were consistently >95% pure using this method (12). T cells were cultured in complete RPMI 1640 (RPMI 1640 [Invitrogen, Carlsbad, CA] supplemented with 10% FBS, 2 mM l-glutamine, 25 mM HEPES, 1 mM sodium pyruvate, 1× nonessential amino acids, 55 nM 2-ME, 100 U/ml penicillin, and 100 μg/ml streptomycin [Mediatech, Manassas, VA]) in the presence of anti-CD3 (145-2C11; BD Pharmingen, San Diego, CA), 4YSyn, or N-4YSyn. Proliferation and inhibition assays were performed, as described (3, 10). MPTP-intoxicated mice received an i.v. tail injection of 5 × 10⁷ freshly isolated SPCs or 1 × 10⁶ freshly enriched Tregs in 0.25 ml HBSS.

CD4⁺ T cells were isolated from N-4YSyn–immunized donors and cultured at 1 × 10⁶/ml with 2 × 10⁶/ml irradiated SPCs and 10 μg/ml N-4YSyn in 20 ml complete RPMI 1640. For polarization, CD4⁺ T cells were cultured for 5 d with 10 ng/ml IL-2 for control Th cells (Thc); 10 ng/ml IL-12 and 2 μg/ml anti–IL-4 for Th1; 10 ng/ml IL-4 and 2 μg/ml anti–IL-12 for Th2; and 3 ng/ml TGF-β, 10 ng/ml IL-6, 5 ng/ml IL-1β, 10 ng/ml IL-23, 2 μg/ml anti–IL-4, 2 μg/ml anti–IL-12, 2 μg/ml anti–IFN-γ, and 2 μg/ml anti–IL-2 (19) for Th17 cells. Each Th subset was harvested, and 10 × 10⁶ T cells from each subset were transferred to separate recipient groups. For stimulation of cytokine production, Th subsets were stimulated with 20 ng/ml PMA and 1 μM ionomycin (Sigma-Aldrich) for 5 h, cells were washed, media were replaced, and supernatants were collected 24 h later for analysis.

Samples from cell fractions were labeled with fluorescently labeled Abs (eBioscience, San Diego, CA) and analyzed by flow cytometry with a FACSCalibur flow cytometer (BD Biosciences, San Jose, CA).

RNA was purified using TRIzol reagent (Invitrogen) and the RNeasy Mini Kit (QIAGEN, Valencia, CA), prior to cDNA synthesis. Real-time PCR analysis using pathway-focused gene-expression profiling arrays (SABiosciences, Frederick, MD) was performed according to the manufacturer’s protocol.

A multianalyte cytokine ELISArray (SA Biosciences) was used for cytokine analysis within cell culture supernatants, according to the manufacturer’s protocol. Absorbance values were read at 450 nm after stopping the reaction. A cytometric bead array for Th1/Th2 cytokines and an IL-17A Flex set (both from BD Biosciences) were used to quantitate cytokine concentrations within culture supernatants. The bead arrays were performed according to the manufacturer’s protocol; the data were acquired on a BD FACSArray bioanalyzer and analyzed using the FCAP Array Software (BD Biosciences). The limits of detection were 0.1 pg/ml for IL-2, 0.03 pg/ml for IL-4, 5.0 pg/ml for IL-5, 1.4 pg/ml for IL-6, 0.5 pg/ml for IFN-γ, 0.9 pg/ml for TNF-α, and 0.8 pg/ml for IL-17A. For in vitro cytokine analyses, CD4⁺ T cells were stimulated with N-4YSyn (10 μg/ml) with or without polarizing culture conditions for 5 d, and cytokine levels were assessed using the mouse intracellular cytokine staining kit (BD Pharmingen). Stimulated CD4⁺ T cells were cultured for an additional 4 h in the presence of PMA, ionomycin, and Brefeldin A (Leukocyte Activation Cocktail, BD Pharmingen), permeabilized, fixed in Cytofix/Cytoperm buffer, and immunostained with PE-conjugated anti–TNF-α, anti–IFN-γ, or anti–IL-17 or -10. Cells were analyzed with an LSRII flow cytometer (BD Biosciences).

Mice were transcardially perfused with PBS, followed by 4% paraformaldehyde (Sigma-Aldrich). Frozen midbrain sections (30 μm) were immunostained for Mac-1 (CD11b, 1:1000; Serotec, Raleigh, NC). Fluorojade C (FJ-C) staining (Millipore, Billerica, MA) was performed on adjacent sections, according to the manufacturer’s protocol, to assess degenerating neurons and was quantified using ImageJ. Overall, dopaminergic neuron survival was assessed 7 d following MPTP intoxication and resolution of cell death processes with polyclonal Abs to mouse tyrosine hydroxylase (TH; 1:1000; EMD Chemicals/Calbiochem, San Diego, CA) and were counterstained for Nissl substance by thionin staining (20), as previously described (21). Total numbers of Mac-1⁺ cells, CD4⁺ T cells, and TH- and Nissl-stained neurons in the SN were estimated by stereological analysis with Stereo Investigator software (MBF Bioscience, Williston, VT), using the optical fractionator module (22). Quantitation of striatal TH (1:500; EMD Chemicals/Calbiochem, Gibbstown, NJ) was performed by densitometric analysis, as described (21). Adjacent midbrain sections were immunostained for CD4 (clone RM4-5, 1:200, BD Pharmingen). Sections were incubated in streptavidin-HRP solution (ABC Elite vector kit, Vector Laboratories, Burlingame, CA) and color developed using a generation system consisting of diaminobenzidine (DAB) chromogen (Sigma-Aldrich), as described (21).

All values are expressed as mean ± SEM. Differences in between-group means were analyzed by one-way ANOVA, followed by the Fisher least significant post hoc test for multiple comparisons (SPSS, Chicago, IL). All effects of treatment were tested at the 95% confidence level.

To confirm and extend our previous findings that N–α-Syn–induced immunocytes exacerbate MPTP-induced inflammation and dopaminergic neurodegeneration, SPCs from donors immunized with N-4YSyn were adoptively transferred to MPTP recipients, and the extent of inflammation and neurodegeneration was determined. Thus, we characterized putative T cell phenotypes and regulatory capacities of CD4⁺ T cells from mice immunized with the nitrated form of the C terminus of α-syn (N-4YSyn), because this portion contains four of five nitratable tyrosine residues. Stereological analysis of Mac-1⁺ cells within the SN 2 d post-MPTP showed >16-fold increase in the number of Mac-1⁺ cells compared with PBS controls (Fig. 1A), whereas the adoptive transfer of N-4YSyn SPC to MPTP recipients increased Mac-1⁺ cell numbers/mm² by 35% of those observed from MPTP treatment alone. FJ-C staining of dead or dying neurons revealed that adoptive transfer of N-4YSyn SPCs accelerated MPTP-induced neuronal death by 7.2-fold (Fig. 1B). Analysis of surviving nigral dopaminergic neurons (TH-immunoreactive [TH⁺] and Nissl-stained [Nissl⁺] neurons) 7 d after MPTP intoxication indicated a 45% overall neuronal loss compared with PBS controls, whereas MPTP-treated recipients receiving N-4YSyn SPCs exhibited a 63% reduction in TH⁺ neurons (Fig. 1C). PBS-treated mice that received N-4YSyn SPCs showed no change in TH⁺ neuron numbers compared with control mice that received PBS alone (data not shown). MPTP mice that received SPCs from PBS/adjuvant- or nonnitrated α-syn (4YSyn)/adjuvant-immunized donors showed no significant additive or protective effect on microglial activation or neuronal survival compared with MPTP alone (data not shown), as previously described (3). No significant effects of any treatment were observed with regard to the numbers of nondopaminergic neurons (TH⁻Nissl⁺). Analysis of CD4⁺ T cell infiltration into the SN revealed that MPTP-intoxicated mice and MPTP-treated recipients of N-4YSyn SPCs had increased infiltration of CD4⁺ cells (8.5- and 10.3-fold, respectively) within the SN at 7 d postintoxication compared with PBS-treated controls (Fig. 1D). These data demonstrate that adaptive immune responses against N–α-syn exacerbate MPTP-induced neuroinflammation and nigrostriatal degeneration and support our previous work (3).

T cells isolated from N-4YSyn–immunized donors stimulated in vitro with anti-CD3 for 24 h produced greater concentrations of IL-17A and TNF-α relative to naive T cells (Fig. 1E). N-4YSyn antigenic stimulation of CD4⁺ Teffs isolated from immunized mice induced IL-17A and -2, TNF-α, and IFN-γ, but not IL-4 or -5 (Fig. 1F), suggesting that immunization partially polarized CD4⁺ T cells in vivo toward a Th1 or Th17 phenotype. Functional characterization of Tregs isolated from immunized FoxP3/GFP transgenic mice revealed that Tregs were functionally deficient in their capacity to inhibit Teff proliferation to anti-CD3 stimulation following immunization with N-4YSyn (20%) compared with Tregs isolated from naive donors (80%) at a ratio of 1:1 (Fig. 1G). Intracellular cytokine tests demonstrated that N-4YSyn–expanded CD4⁺ T cells from N-4YSyn–immunized mice were composed of 38.2% TNF-α–expressing Teffs, 9.0% IFN-γ–expressing Teffs, and 11.2% IL-17A–expressing Teffs (data not shown). Taken together, these results indicate that immunization with N-4YSyn induces Th1 and/or Th17 T cells and produces a deficiency in Treg function. It also suggests roles for Th1 and Th17 in the exacerbation of MPTP-induced inflammation and dopaminergic neuronal death.

To test the abilities of Th1 and Th17 cells to enhance dopaminergic neurodegeneration, CD4⁺ T cells were isolated from N-4YSyn–immunized donors; polarized in vitro for 5 d in culture conditions to favor a Th1, Th2, or Th17 phenotype; and then each polarized T cell subset was adoptively transferred to separate groups of MPTP-intoxicated recipients (Fig. 2A). Cytometric bead array analyses confirmed that culture conditions polarized N-4YSyn CD4⁺ T cells to each designated phenotype, as characterized by the production of IFN-γ by Th1 cells, IL-4 by Th2 cells, and IL-17A by Th17 cells (Fig. 2B). Moreover, flow cytometric analysis of intracellular cytokine expression indicated that polarization of N-4YSyn CD4⁺ T cells under Th1 conditions yielded 4.1% IFN-γ–producing T cells and 59.5% TNF-α–producing T cells. Th17-polarizing conditions yielded 11.8% IL-17A- and 86.8% TNF-α-producing T cells, whereas Th2 culture conditions yielded 69.1% IL-10–producing T cells (data not shown). Compared with nonpolarized CD4⁺ T cells, quantitative PCR analyses of transcription factors and IL genes indicated that polarized Th1 cells showed increased expression of Stat4 (2-fold), Ifng (31-fold), Tnf (2-fold), and Il6 (15-fold); polarized Th2 cells showed increased expression of Gata3 (5-fold), Stat6 (13-fold), Il4 (3-fold), Il10 (5-fold), and Il13 (5-fold); and polarized Th17 cells showed a 351-fold increase in the transcription activator retinoic acid receptor-related orphan receptor c (Rorc) as well as increased expression of Il22 (4-fold), Il23 (8-fold), Il17a (51-fold), and Tnf (76-fold) (data not shown).

Analysis of surviving dopaminergic neurons in TH-immunostained ventral midbrain and striatum 7 d after MPTP treatment and adoptive transfer revealed that adoptive transfer of N-4YSyn Th1 or Th17 cells resulted in decreased numbers of surviving TH⁺ neurons within the SN, whereas only N-4YSyn Th17 cells induced remarkably diminished TH termini densities within the striatum (Fig. 2C). To validate those observations, unbiased stereological analysis of ventral midbrain sections indicated that although PBS-treated controls averaged 7994 ± 212 total TH⁺ neurons within the SN, MPTP intoxication induced a 25% loss of TH⁺ neurons (5971 ± 250) (Fig. 2D). Adoptive transfer of N-4YSyn Th1 cells increased the MPTP lesion to 46% neuronal loss, resulting in 4320 ± 252 total TH⁺ neurons, whereas adoptive transfer of N-4YSyn Th2 cells had no significant protective or exacerbative effect on the total TH⁺ neurons in response to MPTP intoxication. In contrast, adoptive transfer of N-4YSyn Th17 cells induced a 65% loss in the number of surviving TH⁺ neurons (2800 ± 243), a 2.6-fold increased lesion relative to MPTP intoxication alone. Analysis of the surviving TH⁺ dopaminergic neuronal termini showed that the TH density within the striatum of MPTP-intoxicated mice was 26% relative to PBS-treated controls. Adoptive transfer of N-4YSyn Th1 or Th2 cells did not significantly affect the striatal densities relative to MPTP intoxication alone. In contrast, adoptive transfer of N-4YSyn Th17 cells exacerbated the MPTP-induced loss of striatal TH density to 5% of PBS-treated controls (Fig. 2E).

VIP mediates immunosuppression of autoimmunity with increasing Treg numbers or suppressive function (16, 17) or through abrogation of Th17 differentiation (23). To determine whether VIP modulation of N–α-syn–directed immune responses regulate neurodegenerative activities, SPC populations from VIP-treated or N-4YSyn–immunized mice were adoptively transferred separately or together to MPTP recipients; the neuroinflammatory and neurodegenerative responses were evaluated at 2 d post-MPTP intoxication: a time of peak inflammation and rate of neuronal death (24). MPTP mice that received N-4YSyn SPC exhibited an exacerbated inflammatory response, as demonstrated by increased nigral Mac-1 expression, which was diminished in mice treated with VIP SPCs (Fig. 3A). Similarly, neuronal death, as shown by FJ-C staining, was demonstrably increased in N-4YSyn SPC-treated MPTP mice compared with MPTP treatment alone, whereas FJ-C staining was diminished in MPTP mice treated with VIP SPCs and more acutely diminished in mice treated with pooled VIP and N-4YSyn SPCs. Quantitative validation of these observations indicated that transfer of VIP SPCs reduced the numbers of activated microglia by 33% in MPTP-intoxicated mice, whereas transfer of N-4YSyn SPC numbers increased Mac-1⁺ cell densities by 55% (Fig. 3B). More importantly, cotransfer of VIP SPCs with N-4YSyn SPCs attenuated the exacerbative effects mediated by N-4YSyn SPCs by 75% and diminished the numbers of Mac-1⁺ microglia by 61% less than MPTP alone. Analysis of nigral Mac-1⁺ microglia 7 d post-MPTP demonstrated sustained microglial activation in mice that received N-4YSyn SPCs alone or in combination with naive SPCs, whereas significant numbers of Mac-1⁺ cells were not observed within any other treatment group (data not shown). Analysis of total nigral FJ-C⁺ cells revealed a 7-fold increase in the number of dead or dying neurons in MPTP recipients that received N-4YSyn SPCs compared with MPTP alone, whereas the transfer of pooled VIP and N-4YSyn SPCs resulted in 64% fewer injured neurons (Fig. 3C). Adoptive transfer of naive SPCs alone or together with N-4YSyn SPCs showed no significant detrimental or protective effect on microglial activation, and the numbers of FJ-C⁺ neurons were not significantly different compared with MPTP alone (data not shown). These data suggested that a cell population within SPCs from VIP-treated mice, but not naive SPCs was able to inhibit or suppress N-4YSyn–mediated effector cells and ameliorate the exacerbated microglial response and neuropathology.

To validate that neuroprotection was not a transient effect observed only at day 2, we assessed TH-immunostained ventral midbrain and striatal sections 7 d after MPTP treatment and adoptive transfers: a time after which MPTP-induced neuroinflammation and neurodegeneration has subsided (3, 21). MPTP mice that received VIP SPCs showed a modest increase in TH⁺ neuronal density (Fig. 4A). In contrast, TH⁺ neurons within the SN of MPTP recipients that received N-4YSyn SPCs were markedly diminished compared with MPTP treatment alone, whereas those that received pooled VIP and N-4YSyn SPCs exhibited TH⁺ neuronal densities reminiscent of PBS controls. Although lesions of the dopaminergic striatal termini are typically more severe, similar patterns of dopaminergic loss were observed in mice treated with MPTP alone and with separate SPC populations, whereas those treated with pooled N-4YSyn SPCs and VIP SPCs exhibited discernable increases in the density of TH⁺ striatal termini. To confirm these findings, stereological analysis and comparison with PBS controls showed that MPTP reduced SN TH⁺ neurons by 48%, whereas MPTP and N-4YSyn SPCs reduced those neurons by 64% (Fig. 4B). MPTP recipients of VIP SPCs showed a modest, yet significant, 18% increase in TH⁺ neuron number compared with MPTP alone. In comparison, 91% of TH⁺ neurons survived in MPTP mice receiving SPCs from N-4YSyn–immunized and VIP-treated donors. Striatal TH⁺ density in MPTP-intoxicated mice was 16% of PBS controls, whereas the transfer of N-4YSyn SPCs to MPTP recipients reduced striatal TH densities to 7% of PBS controls (Fig. 4C). Although transfer of VIP SPCs to MPTP mice showed no significant additive or protective effect, cotransfer of VIP and N-4YSyn SPCs increased striatal termini survival to 39% of PBS controls. Adoptive transfer of naive SPCs alone showed no significant detrimental or protective effect on dopaminergic neuronal survival (data not shown).

To elucidate the neuroprotective cell populations within the SPC pools, we enriched CD4⁺CD25⁺ Tregs from naive and VIP-treated donors and adoptively transferred each population with N-4YSyn SPCs into MPTP mice. Cotransfer of N-4YSyn and VIP SPCs to MPTP recipients provided 100% protection of TH⁺ nigral dopaminergic neurons; neuroprotection was not observed in MPTP mice that received SPCs from N-4YSyn–immunized and naive donors (45% TH⁺ neuron survival) compared with the percentages of surviving neurons after treatment with MPTP alone (49%) or in combination with N-4YSyn SPCs (34%) (Fig. 4D). In comparison, coadoptive transfer of Tregs from naive or VIP-treated donors with N-4YSyn SPCs afforded significant neuroprotection, with VIP Tregs being more effective (96% survival) than naive Tregs (80% survival). Analysis of striatal dopaminergic termini density was comparable, showing that N-4YSyn SPCs exacerbated the MPTP-induced lesion to 13% of PBS controls (Fig. 4E). Cotransfer of naive with N-4YSyn SPCs was not effective in perturbing N-4YSyn exacerbative effects, whereas cotransfer of VIP and N-4Syn SPCs increased dopaminergic termini survival to 47% of PBS controls, as did cotransfer of naive Tregs and N-4YSyn SPCs (50% of PBS controls). Cotransfer of Tregs from VIP-treated mice with N-4YSyn SPCs was the most efficacious, increasing the mean terminal density to 62% of PBS controls. These results demonstrate that Tregs from VIP-treated mice are most efficacious for neuroprotection of dopaminergic nigral neurons and striatal termini against the neurodegenerative activities of N-4YSyn adaptive immune responses.

Immunization with PBS in adjuvant, N-4YSyn in adjuvant or treatment with VIP afforded little change in the frequencies of splenic CD3⁺, CD19⁺, CD4⁺, and CD4⁺CD25⁺ cells (Fig. 5A). Flow cytometric analysis for CD4⁺CD25⁺ T cells within immune SPC populations revealed minimal increases in the frequencies of CD4⁺CD25⁺ T cells in SPCs of mice immunized with PBS in adjuvant, N-4YSyn in adjuvant, or treated with VIP, whereas the majority of the CD4⁺CD25⁺ T cells also expressed FoxP3⁺. Analysis of Ag-induced proliferative responses showed that although N-4YSyn T cells proliferated in response to N-4YSyn, naive or VIP-treated donors did not [(3) and data not shown]. In contrast, anti-CD3 stimulation of T cells from all donor groups induced proliferative responses in excess of 10-fold. However, such responses were not observed against nonnitrated α-syn in any of the experimental or control groups.

To assess whether VIP SPCs suppress Teff proliferative responses, we evaluated SPC cocultures from N-4YSyn–immunized and VIP-treated donors for their proliferative capacity in the presence of anti-CD3 or N-4YSyn; at a 1:1 ratio of N-4YSyn SPCs/VIP SPCs, proliferation was suppressed by 67% and 81%, respectively, and it diminished in a dose-dependent fashion with the diminution of VIP SPC number (data not shown). Given the dichotomy between Treg and Th17 differentiation, we hypothesized that Treg function or development may be inhibited by immunization with N-4YSyn. To test this hypothesis, we evaluated CD4⁺CD25⁺CD62L^low Tregs isolated from naive, N-4YSyn–immunized, and VIP-treated mice for their capacity to inhibit CD3-mediated proliferation of CD4⁺CD25⁻ naive T cells (Fig. 5B). VIP Tregs showed increased functional capacity to suppress T cell proliferation compared with naive Tregs showing a consistent ≥5% inhibition of proliferation. Importantly, N-4YSyn Tregs were functionally deficient in their suppressive function of Teff proliferation, showing less inhibition of Teff proliferation compared with naive Tregs. In contrast, pooled VIP and N-4YSyn Tregs showed enhanced suppressive capacity compared with all other Treg populations, with 10% greater inhibition versus naive Tregs at a 1:1 Treg/responder ratio. These data suggested that although N-4YSyn immunization abrogates Treg function, VIP Tregs in the presence of N-4YSyn T cells additively restores regulatory function.

To characterize Teff and Treg subset cytokine profiles, isolated CD4⁺ T cells from N-4YSyn–immunized or VIP-treated mice were stimulated with anti-CD3 separately or pooled at a 1:1 ratio to induce cytokine expression. Quantitative RT-PCR revealed that N-4YSyn CD4⁺ T cells showed increased expression of Th17 and Th17-associated genes relative to naive T cells (Fig. 5C). This included Il21, Il17a, and Rorc, whereas genes linked to Th1 (Stat4, Ifng, IL-12, and IL-2), Th2 (Gata3, Stat6, Il4, Il10, and Il13), and Treg (Foxp3 and Il10) (25) were decreased. The increased expression of Il17 and Rorc with concomitant decreases in Th1-, Th2-, and Treg-associated genes suggested that N-4YSyn immunization polarized CD4⁺ T cells toward a Th17 phenotype. Moreover, genes encoding cytokines known to inhibit Th17 differentiation, including IFN-γ, IL-2, IL-4, and IL-15 (−2.85-fold for IL-15; data not shown), were decreased in N-4YSyn T cells compared with naive T cells. Interestingly, VIP T cells showed few changes in gene expression relative to naive T cells; genes predominantly associated with a Th1 phenotype were decreased in expression, whereas the expression of Th2-, Treg-, or Th17-related genes were generally not affected. Importantly, and in contrast, T cells pooled from N-4YSyn–immune and VIP-treated mice showed decreased expression of Th1- and Th17-related genes, whereas genes for Tregs were increased (Fig. 5C, Supplemental Material). Qualitative analysis of cytokine production in response to anti-CD3 stimulation showed increased production of IL-17A and -6 from N-4YSyn T cells relative to naive T cells, whereas the production of IL-2, IFN-γ (Fig. 5D), and IL-4 (data not shown) were decreased. TNF-α production was also increased >2.5-fold relative to naive T cells (data not shown). Analysis of cytokine production by VIP T cells revealed that the cytokine production was not significantly different from that of naive T cells, although the production of Th2-related cytokines IL-10 (Fig. 5D), IL-4, and IL-13 (data not shown) was marginally increased. In comparison, coculture of N-4YSyn and VIP T cells resulted in increased production of regulatory cytokines, including IL-10, IFN-γ (Fig. 5D), and IL-13 (data not shown), with concomitant decreased production of IL-17A and -6 compared with N-4YSyn T cells. TGF-β1 production was increased 2-fold relative to naive T cells in supernatants of N-4YSyn–immunized, VIP-treated, and pooled T cell populations (data not shown). Cytokine concentrations within culture supernatants assessed by cytometric bead array were congruent with these results and revealed that although T cells from N-4YSyn–immunized donors cultured alone produced a nearly 2-fold greater amount of IL-17A relative to naive T cells (87.7 ± 3.3 pg/ml versus 52.2 ± 0.8 pg/ml), coculture with T cells from VIP-treated donors resulted in reduced production of IL-17A, equal to that produced by naive T cells. Production of IFN-γ and TNF-α were also reduced in cultures of pooled N-4YSyn and VIP T cells relative to N-4YSyn T cells cultured alone (data not shown).

We next theorized that VIP could induce antigenic tolerance when given with N-4YSyn immunization. To test this idea, mice treated with or without VIP were immunized with N-4YSyn, and T cells were assessed for their proliferation capacity to N-4YSyn Ag. T cell proliferation to N-4YSyn was suppressed 2.5-fold in T cells isolated from N-4YSyn–immunized and VIP-treated donors compared with T cells from mice immunized with N-4YSyn alone (Fig. 5E, 5F). Notably, T cells from N-4YSyn–immunized mice did not proliferate in response to noncognate, nonnitrated 4YSyn Ags, while demonstrating significant proliferative capacity to cognate N-4YSyn (Fig. 5E). VIP treatment of N-4YSyn–immunized mice reduced 4YSyn-stimulated as well as background T cell proliferation (0–10 μg/ml) compared with non–VIP-treated mice (p < 0.035). VIP treatment of immunized mice also inhibited the capacity to generate any significant N-4YSyn–induced proliferative responses at any Ag concentration compared with media controls (p > 0.05) (Fig. 5F). These data suggest that VIP treatment at the time of N-4YSyn immunization induces Ag-specific tolerance to N-4YSyn. Moreover, T cells from VIP-treated and N-4YSyn–immunized mice showed 1.7- and 1.9-fold increases in the expression of genes encoding Foxp3 and IL-10, respectively, whereas IL-17A gene expression was diminished 2.4-fold compared with T cells from N-4YSyn–immunized mice (Fig. 5G). This demonstrated that VIP treatment favors the induction of regulatory responses that suppress Th17 function.

We present a unique hypothesis for the pathogenesis of PD in that opposing adaptive immune responses mediated by Teffs or Tregs could lead to divergent outcomes in the tempo and progression of disease. As such, the current study was designed to generate a platform for the development of a vaccine strategy that can be applied to intervene against the ongoing inflammatory cascade that occurs in PD. Importantly, the neuroprotective responses seen were distinct from dendritic cell (DC) activities. Simply stated, VIP-induced DCs were not required for the effects observed. Indeed, adoptive transfer of purified populations of Tregs (natural and VIP induced), cells devoid of DCs, attenuated neuroinflammatory and neurodegenerative activities in MPTP-intoxicated and N–α-syn–T cell administered mice. Additional works in models of chronic disease may serve to better translate the findings to human investigations.

Tregs modulate inflammation, attenuate microglial activation, and promote neuronal survival after MPTP intoxication (10, 12, 26, 27). The observation that Treg responses are neuroprotective is contradictory to the hypothesis of protective autoimmunity proposed by other investigators (28–30). Moreover, the T cell repertoire in the elderly is skewed toward a Th2 and Treg phenotype, and an increased suppressive function of CD4⁺CD25⁺ Tregs has been linked to neurodegenerative conditions (31). Possible explanations for this discrepancy are that Tregs have decreased migratory capabilities to the brain in the elderly, or increased inflammation within the brain perturbs Treg function. Another possibility is that the repertoire of naturally occurring Tregs is diminished with age as a result of thymic involution. Our observation that immunization with N-4YSyn induced a Th17 response points to the possibility that increased Tregs may perpetuate Th17 differentiation in the presence of inflammation-associated copious amounts of IL-6 (32). This could occur because Tregs are a primary source of TGF-β. IFN-γ–producing Th1 cells would inhibit Th17 differentiation and support the concept of protective autoimmunity, as observed by other investigators following acute CNS injury (33). Alternatively, Tregs may be deficient in response to neoantigens, such as N–α-syn. Indeed, we showed that Tregs from N-4YSyn–immunized mice were functionally deficient in the suppression of Teff proliferation.

Most notable for the current study is the ability of VIP-induced Tregs to overcome the toxicities of Th17 amplified N–α-syn–mediated nigrostriatal degeneration. Transfer of SPCs from VIP-treated donors alone was less effective than transfer of pooled SPCs from N-4YSyn–immunized donors and VIP-treated donors, suggesting that Tregs need to be activated or Ag-specific to confer neuroprotection. In this study, the systemic inflammation induced by N-4YSyn immunity may have activated Tregs within the VIP population. Alternatively, the phenotype of N-4YSyn CD4⁺ T cells may have been modulated away from a Th17 phenotype in favor of a Treg phenotype mediated by VIP. Our in vitro data lend support to the latter, because the coculture of T cells from N-4YSyn–immunized mice with T cells from VIP-injected mice attenuated the development of a Th17 phenotype in preference of a Treg phenotype. Possible mechanisms include divergent transcriptional regulation in favor of the RORγt/RORc or FoxP3 pathway, because the induction of one activator of transcription may inhibit the induction of the other (32, 34). Alternatively, VIP-induced Tregs could induce greater expression of IL-10 among the Th17 cell population. Increased IL-10 expression was noted in approximately half of TGF-β/IL-6–synergized Th17 cells (35, 36) and was shown to limit Th17-driven inflammation in a mouse model of multiple sclerosis (35). Whether these mechanisms are exclusive or synergistic in our model is still under investigation.

Immunotherapeutic approaches should be directed toward a Th2/Treg response, which promotes Ab production and downregulates proinflammatory Th1/Th17 responses. This is important in light of the observation that T cells in the phase IIa trial of the Aβ vaccination were Th1 based, which underscores the importance of mounting a Th2/Treg-response (37). Methods of generating Tregs are likely to have an important role in the development of therapies for PD, Alzheimer’s disease, stroke, amyotrophic lateral sclerosis, and HIV-1–associated neurocognitive disorders (4, 10–12). Tolerogenic molecules can promote the production of Tregs, including neuropeptides such as VIP, which may be useful as a therapeutic tool to control immunity. Moreover, this work is congruent with our previous findings that immune cells from mice immunized with Copolymer-1 (Copaxone, glatiramer acetate) protect against MPTP toxicities (21). The commonality of results between reports reflects the fact that Copolymer-1 can induce Th2 and Th3 Tregs, and through IL-4 and -10 and TGF-β, it promotes CD4⁺CD25⁻ T cells to CD4⁺CD25⁺ Treg conversions. The interaction of Tregs with Teffs by direct cell contact or through soluble factors may modulate adaptive immune responses, boost attenuation of microglial function, and, thus, provide increased neuroprotection. Taken together, the data provide a platform for a PD immunization strategy. Nonetheless, we realize the limitations seen in the current study. Acute MPTP intoxication does not exactly reflect human disease in course and pathobiology. The chronicity of immune responses also seen in PD is distinct from that observed following acute MPTP exposure. Altogether, the abilities to induce a Treg response, rather than autoreactive immunity toward CNS Ags, may, in fact, generate protective T cell responses without secondarily exacerbating neuronal damage due to inflammation.

We thank Robin Taylor for critical reading of the manuscript and Lisa Kosloski for outstanding technical assistance.

Disclosures The authors have no financial conflicts of interest.