Bee Venom Phospholipase A₂, a Novel Foxp3⁺ Regulatory T Cell Inducer, Protects Dopaminergic Neurons by Modulating Neuroinflammatory Responses in a Mouse Model of Parkinson’s Disease

Among physicians and patients, bee venom (BV) therapy has increasingly gained acceptance for the treatment of immune-related diseases (1–3). Additionally, a number of reports and research articles demonstrating the favorable effect of BV therapy have been published for centuries (4–8). However, the mechanism of its therapeutic immune-modulating effect is poorly understood.

Parkinson’s disease (PD) is a progressive neurodegenerative disease, and neuroinflammation plays a critical role in the pathogenesis of PD (9). Microglia, the primary mediators of neuroinflammation, are innate immune cells of the CNS found in and around degenerating dopaminergic (DA) neurons (10–12). Recently, the importance of the adaptive and innate immune systems in the pathogenesis of PD is becoming increasingly clear (13). Infiltrating T cells are observed around degenerating DA neurons with activated microglia in PD patients and in 1-methyl-4-phenyl-1,2,3,6-tetrahydropyridine (MPTP)–induced mouse models (14–19). Additionally, depletion of CD4⁺ T cells significantly attenuated the DA neuronal death following MPTP injection (16).

Foxp3-expressing CD4⁺ regulatory T cells (Tregs) were introduced as a T cell type that is critical for maintaining immune tolerance under physiological conditions (20). Recently, modulation of Tregs has been proposed as a potential therapeutic approach for neuroinflammation-mediated disorders, including PD (19, 21, 22). We previously observed that BV has a therapeutic effect on MPTP-induced mouse models by increasing the proportion of CD4⁺CD25⁺ Tregs (23).

Tregs may increase in the periphery following the conversion of naive T cells into Tregs through Foxp3 gene expression in response to a variety of stimuli, such as TGF-β, rapamycin, and retinoic acid (24–26). It has been previously shown that PGE₂ converts naive T cells into Foxp3⁺ T cells with a suppressive phenotype (27). Additionally, the conversion of naive T cells into Foxp3⁺ Tregs is directly mediated by regulatory dendritic cells (DCs) that express increased levels of cyclooxygenase (COX)-2, IL-10, TGF-β, or IDO (28, 29).

Bee venom phospholipase A₂ (bvPLA₂) was reported to have high affinity for the mannose receptor (CD206) (30), which is primarily present on the surface of macrophages and DCs (31, 32). The binding of CD206 on DCs activates an immunosuppressive program (33, 34). We hypothesized that bvPLA₂ may have the potential to induce Tregs through CD206⁺ DCs.

This report highlights novel roles for bvPLA₂ in controlling the generation of Tregs and describes its mechanism; these results may contribute to the development of a novel therapeutic agent for regulating the immune system.

All experiments were performed in accordance with the approved animal protocols and guidelines established by the Kyung Hee University [KHUASP(SE)-11-010], and they were conducted on 7- to 8-wk-old male C57BL/6J mice (21–22 g; Japan SLC, Hamamatsu, Japan), Foxp3^EGFP/DTR C57BL/6-transgenic mice (B6.129(Cg)-Foxp3^tm3Ayr/J), Foxp3^EGFP C57BL/6 mice (B6.Cg-Foxp3^tm2(EGFP)Tch/J), and CD206^−/− mice (B6.129P2-Mrc1^tm1Mnz/J) purchased from The Jackson Laboratory (Bar Harbor, ME). All mice were maintained under pathogen-free conditions with air conditioning, a 12/12-h light/dark cycle, and food provided ad libitum.

The splenocytes from C57BL/6J or Foxp3^EGFP mice were resuspended in culture medium and seeded into 48-well plates. The cells were treated with BV (Sigma-Aldrich, St. Louis, MO), melittin (0.5 μg/ml, Sigma-Aldrich), bvPLA₂ (0.1 μg/ml, Sigma-Aldrich), apamin (0.02 μg/ml, Sigma-Aldrich), and mast cell–degranulating peptide (0.01 μg/ml, Sigma-Aldrich), or with various doses of bvPLA₂ in the presence of plate-bound anti-CD3 (5 μg/ml, BD Biosciences, San Jose, CA) and soluble anti-CD28 (2 μg/ml, BD Biosciences) for 72 h. The inactivated bvPLA₂ was obtained by incubating at 100°C for 3 h in a plate heater (35). After the sample data were acquired on a FACScalibur flow cytometer (BD Biosciences), the results were generated in graphical and tabular formats using the FlowJo software (Tree Star, Ashland, OR).

Using magnetic bead separation (CD4⁺CD25⁺ Treg cell kit; Miltenyi Biotec, Bergisch Gladbach, Germany), CD4⁺CD25⁺ and CD4⁺CD25⁻ T cells were isolated from spleens obtained from the Foxp3^EGFP mice that had received PBS or bvPLA₂ as previously described (23). CD4⁺CD25⁻ T cells from PBS-treated mice were labeled with eFluor 670 fluorescent dye (eBioscience, San Diego, CA), plated at a density of 2 × 10⁵ cells/well (responder only) or 1.6 × 10⁵ cells/well in 96-well round-bottom plates and activated with 2 μg/ml soluble anti-CD3 and 6 μg/ml soluble anti-CD28 Ab. CD4⁺CD25⁺ Tregs from PBS- or bvPLA₂-treated mice were added at a density of 0.4 × 10⁵ cells/well. The cultures were incubated for 4 d, and the percentage of mean fluorescence intensity of responder cells was measured. Following the acquisition of sample data on the FACSCalibur flow cytometer (BD Biosciences), the sample results were generated in graphical and tabular formats using FlowJo software (Tree Star).

For MPTP injection, the mice received four i.p. injections of MPTP-HCl (20 mg/kg free base in saline; Sigma-Aldrich) at 2-h intervals, as previously described (23). Twelve hours after the last MPTP injection, the MPTP-injected mice received a single i.p. injection of bvPLA₂ (0.5 mg/kg) or PBS per day for 6 d. Some mice were injected with either bvPLA₂ or vehicle alone as controls. In the case of CD206^−/− mice or Foxp3^EGFP/DTR mice, MPTP-HCl (10 mg/kg free base in saline) was injected i.p. four times a day at 1-h intervals for 2 consecutive days (36). The mortality rate of mice after MPTP injection was ∼0–30% in each group and there are no differences between the groups.

Anti-mouse CD25 rat IgG1 (anti-CD25, clone PC61) Abs were generated in our laboratory from hybridomas obtained from the American Type Culture Collection (Manassas, VA). The mice received a dose of 1 mg/kg anti-CD25 rat IgG1 or normal rat IgG1 daily for 3 d before the MPTP injection. In the case of Foxp3^EGFP/DTR mice, a dose of 20 μg/kg diphtheria toxin or vehicle was injected i.p. every 3 d from 2 d before the MPTP injection. The efficacy of Treg depletion was confirmed by flow cytometry.

Mice were transcardially perfused with a saline solution containing 0.5% sodium nitrate and heparin (10 U/ml) and fixed with 4% paraformaldehyde dissolved in 0.1 M phosphate buffer. The brains were removed, postfixed overnight at 4°C in buffered 4% paraformaldehyde, and stored in a 30% sucrose solution at 4°C until they sank. The frozen brains were sectioned into 30-μm-thick coronal sections using a sliding microtome. All sections were collected in six separate series and processed for immunohistochemical staining. The primary Abs included those directed against tyrosine hydroxylase (TH; 1:2000, Pel-Freez Clinical Systems, Brown Deer, WI), CD11b (1:200, Serotec, Oxford, U.K.), Iba-1 (1:2000, Wako Pure Chemic Industries, Osaka, Japan), ED1 (1:500, Serotec), and CD4 (1:200, Serotec). The stained cells were viewed and analyzed under a brightfield microscope (Nikon, Tokyo, Japan).

The unbiased stereological estimation of the total number of TH⁺ DA neurons, ED1⁺ microglia/macrophages, and CD4⁺ T cells in the substantia nigra (SN) was performed using the optical fractionator method on an Olympus CAST (computer-assisted stereological toolbox system) system, version 2.1.4 (Olympus, Ballerup, Denmark), as previously described (23). The sections used for counting covered the entire SN from the rostral tip of the pars compacta to the caudal end of the pars reticulate (anteroposterior, −2.06 to −4.16 mm from bregma). An actual counting was performed using a 1003 oil objective. The total number of cells was estimated according to the optical fractionator equation (37). More than 300 points were analyzed over all sections of each specimen.

The 6th day after the last MPTP injection, the pole test was performed. In brief, the mice were placed on the top of the gauze-banded wooden pole (50 cm in length and 0.8 cm in diameter) facing upwards and allowed to climb down to the base of the pole. The total time taken for the mouse to reach the floor (locomotion activity time [T-LA]) was recorded, with a cut-off limit of 30 s. Each mouse was given five trials, and the average of the best three measurements was used as the result. Trials with the mouse jumped or slide down the pole were excluded.

Tregs were depleted from Foxp3^EGFP/DTR mice using diphtheria toxin. The Treg-depleted splenocytes were labeled with eFluor 670 fluorescent dye (eBioscience) and plated at a density of 2 × 10⁵ cells in round-bottom plates. The cells were activated with anti-CD3/CD28 in the presence or absence of bvPLA₂ for 18 h, transferred to anti-CD3–uncoated wells to deprive the cells of the anti-CD3 stimuli, and cultured for 2 additional days. After the sample data were acquired on the FACSCalibur flow cytometer (BD Biosciences), the results were generated in graphical and tabular formats using FlowJo software (Tree Star).

The DCs were generated in vitro from the C57BL/6J bone marrow according to the method described by Lutz et al. (38) with some modifications. Briefly, the bone marrow cells from tibiae and femurs were resuspended and cultured. On day 7, the floating cells were collected and purified with magnetic separation beads (CD11c⁺ cell isolation kit; Miltenyi Biotec) to obtain a high-purity population of immature DCs. The percentage of CD11c⁺ cells measured using FACS analysis was >92%. On the same day, the adherent cells were collected as well and analyzed by FACS. More than 95% of these cells were F4/80⁺CD11b⁺ macrophages.

Foxp3⁻CD4⁺ T cells were highly purified through magnetic bead separation (Miltenyi Biotec) from Foxp3⁺ cell–depleted splenocytes obtained from Foxp3^EGFP/DTR mice. The isolated Foxp3⁻CD4⁺ cells were plated at a density of 2 × 10⁵ cells/well in 96-well round-bottom plates and activated with anti-CD3/CD28. The DCs or macrophages were then added at a density of 0.2 × 10⁵ cells/well. In some experiments, the cells were treated with EP antagonists (EP1, ONO8711; EP2, PF04418948; EP3, L798106; EP4, AH23848) or EP2 agonists (AH13205). The cultures were incubated for 3 d, and the percentage of the mean fluorescence intensity of CD4⁺Foxp3⁺ cells was measured. After the sample data were acquired using the FACSCalibur flow cytometer (BD Biosciences), the sample results were generated in graphical and tabular formats using FlowJo software (Tree Star).

To analyze the PLA₂ binding to CD206, we conducted surface plasmon resonance using a Biacore 2000 instrument (Biacore/GE Healthcare Bio-Sciences, Piscataway, NJ). In brief, HBS-EP buffer (GE Healthcare Bio-Sciences) was used for the sample dilution and analysis. The CM5 dextran sensor chip was activated with equal amounts of 0.2 M N-ethyl-N′-(3-diethylaminopropyl)-carbodiimide and 0.05 M N-hydroxysuccinimide. Recombinant human CD206 (5 μg/ml; R&D Systems, Minneapolis, MN) was immobilized in 10 mM sodium acetate buffer (pH 5.5) and then 1 M ethanolamine-hydrochloride (pH 8.0) to deactivate any excess N-hydroxysuccinimide esters. This coupling resulted in 8000 response units of the immobilized proteins per flow cell. To evaluate the binding, each PLA₂ was diluted in HBS-EP buffer, analyzed at various concentrations, and passed over the sensor chip at a flow rate of 20 μl/min. An activated–and-blocked flow cell without immobilized ligand was used to evaluate nonspecific binding. The immobilized surface was regenerated for subsequent measurements using 10 μl 10 mM NaOH. The regeneration solution was passed over the immobilized surface until the surface plasmon resonance signal reached the initial background value before protein injection. For all of the samples, response curves were also recorded on control surfaces. The results were calculated after subtraction of the control values using the BIAevaluation 3.0 software (Biacore).

All values are expressed as the means ± SEM. The statistical significance (p < 0.05 for all analyses) was assessed by one-way ANOVA followed by the Newman–Keuls multiple comparison test for multiple comparisons or by two-tailed Student t test for single comparisons using the Prism 5.01 software (GraphPad Software, San Diego, CA).

To determine which component of BV increases the proportion of Tregs, splenocytes from Foxp3^EGFP mice were stimulated with anti-CD3/CD28 Ab in the presence or absence of BV or its components, including melittin, bvPLA₂, apamin, and mast cell–degranulating peptide. Among the BV components, bvPLA₂ significantly increased the proportion of Tregs within the CD4⁺ T cells; however, the other components of BV did not induce any significant difference in the number of Tregs within this population (Fig. 1A). The ratio of Tregs that increased through bvPLA₂ treatment within the CD4⁺ T cell population depended on the dosage and enzymatic activity of PLA₂. This is because heat–inactivated bvPLA₂ failed to induce Treg differentiation (Fig. 1B). Additionally, the increase in the absolute number of Tregs was dependent on the bvPLA₂ dose (Fig. 1C). Additionally, the effects of bvPLA₂ on the proportion of Tregs in vivo were assessed. In the bvPLA₂-treated mice, the proportion of Tregs was significantly increased in the spleen, compared with that in the PBS-treated control mice, in a dose-dependent manner (Fig. 1D, 1E). However, there was no significant change in the ratio of CD4⁺ T cells, CD8⁺ T cells, and CD19⁺ B cells (Fig. 1F). The suppressive ability of bvPLA₂-treated Tregs was tested through ex vivo proliferation assays using the eFluor 670 fluorescent dye. Tregs from the bvPLA₂-treated mice maintained their functional capacity to suppress responder T cell (CD4⁺CD25⁻) proliferation (Fig. 1G).

To evaluate the effects of bvPLA₂ on MPTP-induced neurotoxicity, mice were administered MPTP and received a single daily i.p. injection of bvPLA₂ (0.5 mg/kg) or saline for 6 d commencing 12 h after the last MPTP injection. On day 7, the brain sections were immunostained with a TH Ab to detect DA neurons. Treatment with bvPLA₂ effectively attenuated the loss of DA neurons in the SN, compared with the MPTP-injected mice (Fig. 2A). Quantification of nigral TH⁺ neurons by stereological counting revealed that bvPLA₂ increased the percentage of TH⁺ neurons by 31%, compared with the mice administered only MPTP (p < 0.05; Fig. 2B). bvPLA₂ alone had no effects on the loss of DA neurons in the SN (data not shown).

To confirm the effect of bvPLA₂ on the behavioral deficits induced by the MPTP injection, mice were treated bvPLA₂ for 6 d after the last MPTP injection, and motor activity was evaluated by the pole test (Fig. 2C). MPTP injections led to significant increase of T-LA compared with saline treatments. In MPTP-treated mice, administration of PLA₂ resulted in a significant decrease of T-LA, indicating improved motor activity. These results show that PLA₂ has neuroprotective effects against MPTP-induced behavioral deficits.

To investigate whether neuroprotection by bvPLA₂ is due to inhibition of MPTP-induced microglial activation in the SN, the brain tissues were processed for immunostaining with a CD11b, Iba-1, and ED1 Abs to detect microglial activation (Fig. 3A–C).

In PBS-injected mice, CD11b⁺ or Iba-1⁺ cells exhibited the typical ramified morphology of resting microglia/macrophages. In contrast, most CD11b or Iba-1⁺ cells displayed an activated morphology, including larger cell bodies with short, thick, or no processed in MPTP-injected mice. Numerous CD11b⁺ or Iba-1⁺ microglia/macrophages were observed in the SN of mice injected with MPTP, dramatically decreased by bvPLA₂ treatment in the MPTP-injected mice (Fig. 3A, 3B). Microglia/macrophages in the MPTP-injected SN appeared to reach a state similar to that of active phagocytes as determined by ED1 immunostaining (Fig. 3C), which labels phagocytic microglia/macrophages in particular, the presence of injured cells or debris. bvPLA₂ reduced the number of ED1⁺ phagocytic microglia/macrophages in the SN of MPTP-injected mice (Fig. 3D).

Previous reports suggest that the infiltration of CD4⁺ T cells into the SN is involved in the DA neuronal loss in MPTP-injected mice by exacerbating an inflammatory process (16). To determine whether bvPLA₂ could modulate the infiltration of CD4⁺ T cells into the SN of MPTP-injected mice, we performed immunostaining with an anti-CD4 Ab (Fig. 3F). bvPLA₂ treatment dramatically reduced the number of CD4⁺ T cells infiltrated into the SN of MPTP-injected mice 7 d after the MPTP injection (75%, p < 0.001; Fig. 3E).

Our previous report showed that Tregs are involved in the therapeutic effect of BV on the MPTP-induced neurotoxicity, which has been verified via Ab-mediated Treg depletion (23). To ascertain whether Tregs were involved in the DA neuronal protection by bvPLA₂, the mice received a PC61 anti-CD25 mAb to deplete Tregs or normal anti-rat IgG1 for an isotype control for 3 d before MPTP injection. The MPTP-induced TH⁺ neuronal loss did not differ between the Treg-depleted and control mice; however, the neuroprotective effect of bvPLA₂ was reversed in the Treg-depleted mice (p < 0.05; Fig. 4A, 4B). Even though PC61 depletes Tregs, it also depletes CD25-expressing CD4⁺ effector T cells as well. To exclude the non-Treg–related effect from the neuroprotective effect by bvPLA₂, we selectively depleted Foxp3⁺ Tregs with diphtheria toxin treatment in Foxp3^EGFP/DTR mice before MTPT injection. The neuroprotective effect of bvPLA₂ vanished in the Treg-depleted mice (p < 0.05; Fig. 4C). The data suggest that the neuroprotective effect of bvPLA₂ is associated with the modulation of the peripheral immune tolerance of Tregs in the MPTP mouse model.

To confirm that bvPLA₂ induces Treg differentiation, the existing Tregs were selectively depleted in the Foxp3^EGFP/DTR mice using diphtheria toxin, and then induced Tregs from the Treg-depleted splenocytes de novo. The results showed that bvPLA₂ induced a 4-fold higher Treg differentiation compared with PBS in the splenocytes (Fig. 5A).

A previous study reported that bvPLA₂ has a high affinity with the mannose receptor (CD206) (30), which is primarily present on the surface of macrophages and DCs (31, 32).

To evaluate the necessity of CD206 for bvPLA₂ binding in DCs, DCs from WT or CD206 knockout mice were treated with bvPLA₂ preconjugated with Alexa Fluor 488. The population of CD206⁺ DCs from WT mice was 34.2 ± 4.1%, as determined by flow cytometry, and they were bound with bvPLA₂–Alexa Fluor 488, but the CD206⁻ DCs were not bound with bvPLA₂–Alexa Fluor 488 (Fig. 5B). To confirm the binding of bvPLA₂ to CD206, surface plasmon resonance analysis was performed. bvPLA₂ was applied to a recombinant human CD206-immobilized CM5 dextran sensor chip. The sensograms showed that bvPLA₂ has binding affinity for the CD206 protein in a concentration-dependent manner (K_D = 4.79 × 10⁻⁶ M, Fig. 5C).

Because it has been reported that the binding of CD206 on DCs activates an immunosuppressive program (33, 34), it is necessary to identify a specific cell subset that shows an immune-modulating effect. Next, we checked whether bvPLA₂ had an effect on Foxp3⁻CD4⁺ T cells directly or other cells required for Treg differentiation. bvPLA₂ could not induce Tregs in cultures of Foxp3⁻CD4⁺ T cells alone or in cocultures of Foxp3⁻CD4⁺ T cells with macrophages; however, it significantly induced Tregs in coculture of Foxp3⁻CD4⁺ T cells with DCs (Fig. 5D). This result indicates that DCs are required for Treg differentiation by bvPLA₂. To clarify that CD206 is a receptor for bvPLA₂ that results in an increase in Tregs, we cultured highly purified Foxp3⁻CD4⁺ T cells with DCs from CD206^+/+, CD206^+/−, or CD206^−/− mice in the presence or absence of bvPLA₂. As expected, Tregs were induced by bvPLA₂ treatment in cocultures of Foxp3⁻CD4⁺ T cells with DCs from CD206^+/+ or CD206^+/− mice but not DCs from CD206^−/− mice or macrophages from CD206^+/+, CD206^+/−, or CD206^−/− mice (Fig. 5E). Additionally, to ascertain whether the increase in DA neurons caused by bvPLA₂ treatment was directly associated with existence of CD206, neuroprotective effects of bvPLA₂ was accessed in MPTP-injected CD206^−/− mice. The results showed that MPTP-induced TH⁺ neuronal loss did not recover upon bvPLA₂ treatment in MPTP-injected CD206^−/−mice (Fig. 5F).

To understand how bvPLA₂ induces Treg differentiation through DCs, we analyzed the mRNA profile of DCs treated with bvPLA₂ or PBS. bvPLA₂-treated DCs expressed higher levels of COX-2 mRNA compared with the control cells; however, the levels of the other mRNAs, including TGF-β, IL-10, and IDO, were not altered by bvPLA₂ treatment (Fig. 6A). To confirm that the enhanced COX-2 expression results in PGE₂ secretion, the resultant culture SN were tested for PGE₂ release using ELISA. The amount of PGE₂ was directly proportional to the dose of bvPLA₂ (Fig. 6B). Moreover, NS-398, a COX-2 inhibitor, completely inhibited the effect of bvPLA₂ on Treg induction (Fig. 6C). DCs from CD206^+/+ mice significantly increased the amount of PGE₂ secreted in response to bvPLA₂, whereas DCs from CD206^−/− mice or macrophages from both CD206^+/+ and CD206^−/− mice did not respond to bvPLA₂ (Fig. 6D). These data indicate that PGE₂ from CD206⁺ DCs induces Treg differentiation.

There are four G protein–coupled receptors that respond to PGE₂, and these are designated as subtypes EP1, EP2, EP3, and EP4. The EP subtypes exhibit differences in signal transduction, and the PGE₂ actions mediated by each subtype identify the role that each EP subtype plays in various physiological and pathophysiological responses (39). We treated the mice with EP antagonists to determine the type of EP receptor that is associated with Treg induction via bvPLA₂. The results showed that the EP2 antagonist reduced the effect of bvPLA₂; however, the other EP antagonists did not alter the effect of bvPLA₂ (Fig. 6E). We confirmed that PGE₂ and the EP2 agonist directly affect the increase in Foxp3 expression in Foxp3⁻CD4⁺ T cells (Fig. 6F). These data indicate that PGE₂ secreted from DCs increases Treg differentiation through the EP2 receptor in Foxp3⁻CD4⁺ T cells.

In the present study, to our knowledge, we first demonstrated that bvPLA₂ is the major BV compound capable of inducing Treg expansion without altering the total composition of the other cell types in vivo and in vitro. Our previous report showed that BV has therapeutic effects on the MPTP-induced mouse model of PD via modulating the neuroinflammatory response and increasing the proportion of functional Tregs (23). In this study, we showed that bvPLA₂ has neuroprotective effects by suppressing microglial activation and reducing the infiltrating CD4⁺ T cells in the MPTP-induced mouse model of PD. In addition to the neuroprotective effects, bvPLA₂ directly binds to the mannose receptor (CD206) on DCs, and this binding induces the release of PGE₂, which promotes Treg induction in CD4 T cells.

Microglia are resident innate immune cells of the CNS found in and around degenerating neurons (10–12) that are rapidly activated in response to neuronal damage and significantly contribute to secondary neuronal loss (40, 41). MPTP, a toxin that can trigger a severe and irreversible PD-like syndrome, induces selective DA neuronal death followed by a robust and sustained inflammatory reaction that causes secondary DA neuronal loss in the SN (17, 42, 43). Several reports show that BV inhibits microglial activation, followed by a reduction of secondary neuronal loss in animal models of neurodegenerative diseases, including PD (23, 44, 45). Although investigations on BV began almost 100 y ago, its mechanism of anti-inflammation is still uncertain. BV is a complex mixture of peptides and proteins such as melittin, PLA₂, and apamin (46, 47). In the present study, we showed that among the BV compounds, bvPLA₂ protects DA neurons against MPTP intoxication by attenuating microglial activation in the MPTP mouse model of PD.

Our previous study suggested that increasing Tregs contributes to the neuroprotective effects of BV in the MPTP mouse model of PD (23). Recently, it has become evident that the adaptive immune system is of importance in PD pathogenesis and progression by recruiting T cells to the SN (13). Specifically, the adoptive transfer of T cells from mice immunized with N-α-synuclein led to a robust neuroinflammatory response with accelerated DA neuronal death in the SN of MPTP-injected mice (48). Additionally, it was demonstrated that CD4⁺ T cells contribute to DA neurodegeneration through the Fas/FasL pathway in an MPTP-induced mouse model of PD (16). Several studies suggested that CD4⁺ effector T cells, such as Th1 and Th17, have been considered potential contributors to neuroinflammation (19). Consistent with these reports, our results show that microglial activation and infiltration of CD4⁺ T cells are reduced by bvPLA₂ treatment in MPTP-injected mice (Fig. 2). These results suggest that the neuroprotective effects of bvPLA₂ are associated with reduced infiltration of CD4⁺ effector T cells into the SN of MPTP-injected mice.

Tregs were found to play a pivotal role in controlling the immune response of peripheral CD4⁺ T cells and, therefore, in maintaining tolerance under physiological conditions (20). Tregs have neuroprotective effects by suppressing microglia or effector T cells verified in two independent experimental paradigms, Ab-mediated Treg depletion (21) and adoptive transfer of Tregs (22). In the present study, we showed that bvPLA₂ was not neuroprotective in mice depleted of Tregs (Fig. 4). These results indicated the possible involvement of Tregs in the effects of bvPLA₂ in attenuating microglial responses and in protecting against MPTP-induced DA neuronal damage. These findings have suggested that promotion of the expansion or induction of Tregs may have therapeutic potential for treating PD. In particular, modulation of Tregs has been proposed as a potential therapeutic approach for various neurodegenerative diseases, including PD, multiple sclerosis, and amyotrophic lateral sclerosis (21, 22).

bvPLA₂ was previously reported to have a high affinity for the mannose receptor in macrophages, which is known as CD206 (30). CD206 is a C-type lectin carbohydrate-binding protein that is primarily present on the surface of macrophages and DCs (31, 32). In this study, bvPLA₂ binding to the CD206 receptor was firmly confirmed by surface plasmon resonance assay (Fig. 5C), and the necessity of CD206 receptor expression in DCs for Treg differentiation was confirmed by the observation that CD206^−/− DCs lost their Treg-inducing role (Fig. 5D). Interestingly, only a subset of DCs expresses CD206 (Fig. 4B) and CD206⁺ macrophages did not affect Treg induction by bvPLA₂ (Fig. 4D). The possible uptake of bvPLA₂ by DCs through CD206 and further induction of COX-2 expression may lead Treg differentiation. Additionally, the inhibition of PGE₂ synthesis by a COX-2 inhibitor prevented bvPLA₂-mediated Treg induction (Fig. 6C). These results are justified by previous observations that mannose receptor signaling promotes Treg induction through the COX-2/PGE₂ pathway in vitro (34). Additionally, it is well documented that PGE₂ upregulates Foxp3 at both the mRNA and the protein levels and enhances Foxp3 promoter activity (27), and that the PGE₂-mediated induction of Foxp3 gene expression in Tregs is ablated in the absence of EP2 receptor expression on CD4⁺ T cells (49). In our study, PGE₂ production induced by bvPLA₂-stimulated DCs exerts an immunologic effect through the EP2 receptor on CD4⁺ T cells. The Treg-inducing effect of bvPLA₂ is not directly exerted on T cells. To elucidate the detailed function of bvPLA₂, further mechanistic studies are required.

It is well known that PLA₂ catalyzes the hydrolysis of the sn-2 fatty acyl bond of phospholipids to liberate free fatty acids and lysophospholipids. This enzyme, which is found in the venom of various species, including reptiles and insects, is classified into different groups depending on its chemical characteristics, and it mediates a large spectrum of toxic effects (5). Recently, it was demonstrated that selectively secreted endogenous PLA₂-IID from Tregs promoted the differentiation of Tregs, presumably via the PI3K/Akt/mammalian target of rapamycin pathway, and suppressed proliferation of CD4⁺ and CD8⁺ T cells (50). It has been also demonstrated that bvPLA₂ induces Th2 response, which may have contributed to the suppression in the experimental model of PD (35, 51). Additionally, an enhanced level of PGE₂ by bvPLA₂ treatment may promote the Th2 response (52). These findings suggest that there could be alternative pathways of bvPLA₂-mediated protection on PD besides the mechanism by which bvPLA₂ promotes Treg differentiation via PGE₂ induction.

In conclusion, our findings demonstrate a novel effect of bvPLA₂: the mediation of immune suppression through controlling PGE₂ secretion by CD206⁺ DCs. PGE₂ released by bvPLA₂-stimulated DCs acts on the EP2 receptor in CD4⁺ T cells, and this binding leads to the upregulation of Foxp3 expression. bvPLA₂ may thus be a potential candidate for treating PD and other disorders associated with neuroinflammation.

The authors have no financial conflicts of interest.