Abstract
Cytosolic phospholipase A2α (cPLA2α) is the rate-limiting enzyme for release of arachidonic acid, which is converted primarily to PGs via the cyclooxygenase 1 and 2 pathways and to leukotrienes via the 5-lipoxygenase pathway. We used adoptive transfer and relapsing–remitting forms of experimental autoimmune encephalomyelitis (EAE), an animal model of multiple sclerosis, in two different strains of mice (SJL or C57BL/6) to demonstrate that blockade of cPLA2α with a highly specific small-molecule inhibitor during the tissue-damage effector phase abrogates the clinical manifestation of disease. Using the adoptive transfer model in SJL mice, we demonstrated that the blockade of cPLA2α during the effector phase of disease was more efficacious in ameliorating the disease pathogenesis than the blockade of each of the downstream enzymes, cyclooxygenase-1/2 and 5-lipooxygenase. Similarly, blockade of cPLA2α was highly efficacious in ameliorating disease pathogenesis during the effector phase of EAE in the adoptive transfer model of EAE in C57BL/6 mice. Investigation of the mechanism of action indicates that cPLA2α inhibitors act on APCs to diminish their ability to induce Ag-specific effector T cell proliferation and proinflammatory cytokine production. Furthermore, cPLA2α inhibitors may prevent activation of CNS-resident microglia and may increase oligodendrocyte survival. Finally, in a relapsing–remitting model of EAE in SJL mice, therapeutic administration of a cPLA2α inhibitor, starting from the peak of disease or during remission, completely protected the mice from subsequent relapses.
Experimental autoimmune encephalomyelitis (EAE) (1) is a T cell-mediated inflammatory disease of the CNS that clinically manifests as ascending paralysis. EAE shares many clinical and histopathological features with multiple sclerosis (MS) and is a commonly used animal model of this human autoimmune disease (1–3). Immunization with CNS myelin proteins or peptides or adoptive transfer of myelin-specific CD4+ T cells induces EAE in susceptible animals. Clinically distinct EAE develops in different mouse strains, each of which is believed to represent different aspects of MS. For example, chronic-progressive EAE develops in C57BL/6 (B6) mice immunized with myelin oligodendrocyte glycoprotein (MOG)35–55 peptide, whereas a remitting–relapsing form of EAE develops in SJL mice immunized with proteolipid peptide (PLP)139–151 (4). In addition, encephalitogenic myelin-specific T cells from immunized donors can be adoptively transferred to syngeneic naive recipient mice to induce EAE. Agents that are efficacious in preventing EAE by prophylactic treatment have had limited or no success in treatment of human MS (4–7). However, agents that generally showed efficacy by therapeutic dosing, especially in more than one EAE model with several different mouse strains, have been relatively more successful in treating MS (4, 8). For example, although anti-p40 Abs that block IL-12/IL-23 have been highly efficacious in various models of EAE in rodents (9–11) as well as in primates (12, 13), they met with failure in human clinical trials of MS (14–16). A closer look at the different EAE models used for testing the anti-p40 Abs indicated that most of these models were using a prophylactic dosing regimen, with the Abs being administered early during the induction phase of the disease, when encephalitogenic effector cells were being generated in the peripheral compartment. To study the role of these cytokines during the induction and effector phases of EAE separately, we reported a robust adoptive transfer model of EAE in B6 mice, in which fully differentiated encephalitogenic effector T cells were adoptively transferred into naive syngeneic recipients to induce reproducibly strong EAE (17). Using this model, we demonstrated that therapeutic dosing with anti-p40 Abs to block IL-12/IL-23 during the tissue-damage effector phase was not efficacious in protecting mice from EAE (17). Thus, collectively these results suggest that simply blocking IL-12/IL-23 during EAE or during the ongoing clinical episode of MS may not be sufficient for amelioration of disease symptoms. Indeed, in addition to the Th1-type and Th17-type cytokines, many other inflammatory mediators are increased during different phases of EAE and MS, and it is likely that some of these may also be critical for the pathogenesis of EAE and MS.
Cytosolic phospholipase A2α (cPLA2α) is one of the critical enzymes involved in generation of multiple proinflammatory mediators, many of which have been implicated in the pathogenesis of MS and EAE (18–21). cPLA2α selectively cleaves arachidonyl phospholipids to release free arachidonic acid (AA), which is converted primarily to PGs via the cyclooxygenase (COX)-1/2 pathway, and leukotrienes [LTs; such as leukotriene B4 (LTB4) and cysteinyl LTs] via the 5-lipoxygenase (5-LO) pathway. Concurrent with the generation of AA, cPLA2α also releases lysophosphatidylcholine (LPC), which serves as the precursor for platelet-activating factor (PAF) (22, 23). All three classes of lipid mediators are increased in cerebrospinal fluid of MS patients (18–21), and there is evidence suggesting that they play a pathogenic role in EAE development (24–28). We reported that mice deficient in cPLA2α (cPLA2α−/−), which are deficient in production of PGs, LTs, and PAF, are resistant to EAE induction (24). In addition, MOG-specific T cells generated from cPLA2α−/− donors and adoptively transferred to wild-type (WT) recipients or generated from WT donors and adoptively transferred to cPLA2α−/− recipients induced less severe EAE in the recipient mice in each case, indicating a role for cPLA2α during both the induction and the effector phases of the disease (24). However, it is possible that cPLA2α is essential for development of the immune system but not critical for immune responses when adult animals are challenged. To address this, we recently developed highly selective small-molecule cPLA2α inhibitors—ecopladib, efipladib, and WAY-196025—that belong to the indole class of compounds (29–31). These small-molecule inhibitors are highly selective for cPLA2α over the closely related cPLA2-β, -γ, and -ζ isoforms, and block production of PGs, LTs, and PAF in whole-blood assays. Using one of these indole-based compounds, WAY-196025, we recently confirmed the role of cPLA2α in EAE (32). Prophylactic dosing limited to the duration of disease induction demonstrated that blocking cPLA2α during the induction phase of EAE prevented EAE development and greatly reduced Ag-induced production of Th1-type cytokines and IL-17 (32). This is in agreement with recent reports suggesting that PGs, especially PGE2, may act directly on both human and murine T cells to facilitate their conversion to Th17 phenotype and effector cytokine production (33, 34). Thus, there is ample evidence to suggest that blocking cPLA2α in inflammatory and autoimmune diseases, including those mediated by Th1 and Th17 cells, would be beneficial because it would block the production of multiple cytokine and noncytokine proinflammatory mediators. However, it is important to note that in addition to being converted into proinflammatory mediators, AA is also a precursor of anti-inflammatory lipid mediators such as lipoxins and agonists of peroxisome proliferator-activated receptor γ. These anti-inflammatory mediators have been shown to enhance resolution of inflammation and reduce severity of EAE (35–37) by negatively regulating both macrophage and T cell functions (38). Because of this interplay of pro- and anti-inflammatory arms of the cPLA2 pathway, the outcome of cPLA2α blockade during the CNS tissue-damage effector phase of EAE, which more closely represents the CNS pathological phase of MS, is not clear. Furthermore, although prophylactic dosing with cPLA2α inhibitors during the induction phase was efficacious in preventing EAE (32), it may not result in clinical efficacy in MS. To understand the mechanism of cPLA2α during the various phases of the disease and to determine the outcome of cPLA2α blockade during the CNS tissue-damage effector phase of EAE, we used the adoptive transfer and relapsing–remitting models of EAE together with therapeutic blockade of cPLA2α with WAY-196025. We also compared the efficacy of cPLA2α blockade with that of individual blockade of downstream enzymes, COX-1 and -2 and 5-LO, to understand their relative roles in the pathogenesis of the effector phase of EAE.
In the current study, we demonstrate that blockade of cPLA2α during the tissue-damage effector phase of EAE using a highly specific small-molecule inhibitor, WAY-196025, completely blocked EAE development in the adoptive transfer models in both SJL and B6 mice. Blocking COX-1/2 reduced severity of EAE, whereas blocking 5-LO delayed onset but did not affect the maximum severity of EAE. In the relapsing–remitting EAE model in SJL mice, therapeutic administration of cPLA2α inhibitors, starting from the peak of disease or during the remission, completely protected the mice from subsequent relapses of EAE during the course of treatment. In addition, an investigation of the mechanism of action indicated that cPLA2α inhibitors act on APCs to diminish their ability to induce Ag-specific effector T cell proliferation or cytokine production and at the same time enhance their ability to induce Ag-specific regulatory T cell (Treg) differentiation. These results indicate that cPLA2α plays a critical pathophysiological role during the tissue-damage effector phase of EAE, in part by its action on APCs, and is a potential target for the treatment of MS.
Materials and Methods
Mice
Female C57BL/6, SJL, and B6.PL (Thy1.1+) mice were purchased from The Jackson Laboratory or Taconic Laboratory. Anti-myelin basic protein (MBP) TCR transgenic mice in the B10.PL background have been previously described (39) and were bred at our contract laboratory (Taconic). All mice were used at 6–12 wk of age. The Institutional Animal Care and Use Committee at Wyeth Research approved all protocols.
EAE induction
The direct immunization model has been described previously (32). Briefly, EAE was induced in SJL mice using immunization with PLP139–151. The mice were injected s.c. at two sites dorsally with a total of 200 μg PLP139–151 in CFA containing 6 mg/ml killed Mycobacterium tuberculosis.
Preparation of cells for EAE induction in adoptive transfer model
Preparation of MOG35–55-specific cells for adoptive-transfer EAE mice has been described previously (17). Briefly, splenocytes were harvested from MOG-immunized donors (B6 or B6.PL) on day 11 and were cultured in the presence of MOG35–55 with recombinant murine IL-12 (30 ng/ml) and anti-murine IFN-γ (clone XMG1.2; 10 μg/ml; BD Pharmingen) to generate encephalitogenic cells. Three days after initiation of the cultures, cells were harvested, washed, and 10 × 106 to 15 × 106 encephalitogenic cells injected into recipient mice i.p.
Preparation of PLP139–151-specific cells for adoptive-transfer EAE mice has been described previously (9). Briefly, splenocytes were harvested from PLP-immunized donors on day 11 and were cultured in the presence of PLP139–151 in the presence or absence of recombinant murine IL-12 (30 ng/ml) or recombinant murine IL-23 (10 ng/ml). Three days after initiation of the cultures, cells were harvested, washed, and 15 × 106 to 25 × 106 encephalitogenic cells injected into recipient mice i.p.
To test if the treatment can prevent EAE development in the adoptive transfer model, SJL recipient mice were treated with vehicle, cPLA2α (WAY-196025, 50 mg/kg/dose), COX-1/2 (naproxen, 50 mg/kg/dose), or 5-LO (zileuton, 50 mg/kg/dose) inhibitors s.c., twice per day, starting on the day of adoptive transfer. For adoptive transfer experiments in B6 (or B6.PL) mice, the recipient mice were treated with vehicle or WAY-196025 at either 100 mg/kg/dose s.c., once per day, or 50 mg/kg/dose s.c., twice per day, starting on the day of adoptive transfer. To test if the treatment can alter the course of existing clinical EAE and prevent relapses in the direct immunization model, treatment of SJL mice was initiated on day 12 (peak of clinical signs of EAE), or after the first episode of EAE when the mice were randomized into vehicle or WAY-196025 treatment groups. The aqueous vehicle (Phosal) contained 15.7% Phosal 53 MCT (Phospholipid GmbH, Köln, Germany), 4.71% propylene carbonate (Alfa Aesar, Ward Hill, MA), 4.71% Labrasol EP (Gattefosse, Saint Priest, France), and 1.58% polysorbate 80 NF (Mallinckrodt Baker, Phillipsburg, NJ). Disease severity, scored in a blinded fashion, was expressed as a mean of daily EAE scores of all the mice in the group. In addition, mean maximum severity (MMS) score was calculated for each group by using the highest daily EAE score of each mouse in the group during the course of the experiment. Finally, cumulative disease severity was expressed as area under the curve (AUC), which was calculated for each mouse by adding all the daily scores during the course of the experiment. Paralysis (clinical evidence of EAE) was assessed daily, starting on the day of immunizations or adoptive transfer, when all the mice were still clinically normal. Clinically, animals were scored as follows: 1, limp tail; 2, partial hind leg paralysis; 3, complete hind leg paralysis or partial hind and front leg paralysis; 4, complete hind and partial front leg paralysis; 5, complete hind and partial front leg paralysis and reduced responsiveness to external stimuli. Mice were euthanized immediately if they scored 5 or if they scored 4 two days in a row.
Histology
CNS tissue preparation for histological evaluation was done as previously described (32). Briefly, mice were euthanized using CO2 asphyxia and immediately perfused with 10–20 ml PBS. The head and spine were dissected in one piece, and a hole was made in the skull over the cisterna magna to allow formalin to enter the subarachnoid space. The tissue was immersed into 10% buffered formalin and fixed for at least 72 h before processing. After fixation, brains were removed from the skull prior to sectioning. Spinal columns with spinal cords were decalcified prior to sectioning in situ. Four transverse sections of the brain (frontal cortex to medulla) and multiple (n = 9–13 sections/animal) transverse sections of the cervical, thoracic, and lumbar spinal cord were routinely prepared and stained with H&E. The numbers of inflammatory foci containing at least 20 cells were counted in each H&E-stained section in a blinded fashion by the same pathologist (N. Stedman). When foci were coalescing, estimates were made of the number of foci. The presence of vacuolation and pallor in the white matter were also noted. Demyelination was assessed on Luxol fast blue sections.
For immunohistochemical analysis of CNS, the following protocol was used. After treatment with vehicle or WAY-196025, brains and spinal cords were removed, fixed in 4% paraformaldehyde at 37°C for 24 h, and then transferred to 30% sucrose. Representative coronal sections (25 μm) were obtained with the help of a rotary microtome (RM2145; Leica, Nussloch, Germany). Coronal sections of 25 μm were taken for staining. The sections were incubated overnight at 4°C with one of the following Abs: either an anti-CD11 (activated microglia) mAb (1:200; Cedarlane Canada) or an anti-Oligodendrocyte +04 polyclonal Ab (1:500; Chemicon). After washing, a fluorescent-conjugated secondary Ab (1:2000, Alexa Fluor 488 or 594; Molecular Probes, Eugene, OR) was applied for 2 h at room temperature. Fluorescent images were viewed and collected under an Axiovert 135 fluorescence microscope (Carl Zeiss, Jena, Germany). The number of activated microglial cells in the cortex (as recognized by positive OX-42) and Oligodendrocyte +04–positive cells staining in the thoracic spinal cord was evaluated by counting five fields/section, six sections/animal, five animals/condition using a 40× objective on an Axiovert 135 fluorescence microscope (Carl Zeiss). Images were captured using Axiovision 4.5 software. The ratio of the number of activated microglia (OX-42) or Oligodendrocyte +04–positive cells to the total number of vehicle per field was calculated for each section.
Intracellular cytokine staining to characterize cells from B6.PL (Thy1.1+) donors in CNS of recipient mice
As described earlier for B6 adoptive transfer experiments, splenocytes were harvested from MOG35–55-immunized WT B6.PL (Thy1.1+) donor mice on day 11 and were cultured with MOG, IL-12, and anti–IFN-γ for 3 d and adoptively transferred into WT B6 (Thy1.2+) recipients. At peak disease (11 d after adoptive transfer), CNS tissue was harvested from PBS-perfused recipient mice, single-cell suspension was prepared, washed, and resuspended in 70% Percoll (Sigma-Aldrich), overlaid with 37% Percoll, and centrifuged at 600 × g for 25 min with no brake. CNS mononuclear cells were obtained from the interface of 37/70% Percoll gradient, washed, and cultured with MOG for 24 h. During the last 4 h of culture, PMA (50 ng/ml; Sigma-Aldrich), ionomycin (500 ng/ml; Sigma-Aldrich), and GolgiPlug were added at the manufacturer’s recommended concentration (BD Pharmingen). Cells were first stained extracellularly with different fluorochrome-conjugated Abs [Thy1.1 (OX-7) FITC, Ms IgG1κ FITC; CD4 (RM4-5) PE–Cy7, rat IgG2aκ PE–Cy7; CD45RB (16A) PE, rat IgG2aκ PE; CD44 (IM7) allophycocyanin, rat IgG2bκ allophycocyanin), fixed and permeabilized with Cytofix/Cytoperm solution (BD Pharmingen), and were then stained intracellularly with Alexa 647-conjugated anti–IFN-γ (XMG1.2) and PE-conjugated anti–IL-17A (TC11-18H10). A minimum of 2 × 106 events per group was acquired on either FACSCalibur (BD Biosciences) or CYAN (DakoCytomation), and data were analyzed with CellQuest-Pro software (BD Biosciences) or Summit (DakoCytomation). For the purposes of setting gates, appropriate fluorochrome-conjugated isotype controls for each of the surface- and intracellular-staining Abs were used. All fluorochrome-conjugated Abs and the corresponding isotype controls were purchased from BD Pharmingen.
Crisscross experiment
In these experiments, the Ag-experienced APCs and naive CD4+ T cells came from different donor mice. B10.PL mice were immunized with MBP1–19 and were treated with vehicle or WAY-196025 100 mg/kg, s.c., once per day for 8 d, and their spleens were isolated. APCs were isolated from these vehicle-treated or WAY-196025–treated B10.PL mice by positive selection magnetic beads to remove CD4+ and CD8+ T cells, using the manufacturer’s protocol (Invitrogen Life Technologies). The leftover cells (CD4− CD8−) were used as APCs for setting up the proliferation cultures with CD4+ T cells or immediately analyzed by flow cytometry. For CD4+ T cells, we harvested lymph node and spleens from naive anti-MBP TCR transgenic mice and prepared single-cell suspensions. Naive anti-MBP TCR transgenic CD4+ T cells were isolated by negative selection using magnetic sorting, according to the manufacturer’s protocol (Invitrogen Life Technologies). For proliferation assay, APCs were irradiated at 2000 rads and cultured with naive anti-MBP TCR transgenic CD4+ T cells in a ratio of 5:1 in presence of various amounts of MBP1–19 peptide with or without TGF-β (10 ng/ml). Forty-four to forty-eight hours after the initiation of cultures, supernatant was harvested, and the cultures were pulsed with 0.5 μCi [3H]thymidine and harvested 14–18 h later to determine proliferation. Concentrations of IL-10, IFN-γ, and TNF in the supernatants were quantified using a cytometric bead array kit from BD Pharmingen.
Statistical analysis
For statistical analysis, a Poisson distribution was used to model the inflammatory foci parameter. A square root transformation was applied to stabilize the variance, and then the transformed data were analyzed with a one-way ANOVA. For histological analysis, severity scores were analyzed using the mean score Mantel–Haenszel statistic. Clinical scores were compared using ANOVA.
Results
Therapeutic dosing with cPLA2α inhibitor but not COX-1/2 inhibitor or 5-LO inhibitor abrogates EAE in the adoptive transfer model in SJL recipients
We previously demonstrated that cPLA2α−/− mice are resistant to EAE (24). Furthermore, we demonstrated that prophylactic treatment with a cPLA2α-specific small-molecule inhibitor, WAY-196025 (30), prevented EAE development in WT mice (32). To test whether the cPLA2α inhibitor would be efficacious in the tissue-damage effector phase of EAE, which is more relevant to MS pathology (14, 17), we used adoptive transfer models of EAE where the induction phase and the effector phase of EAE can be studied separately. Splenocytes from PLP139–151-immunized SJL donor mice were cultured in vitro with PLP139–151 to generate PLP-specific encephalitogenic cells. These cells were then adoptively transferred into naive syngeneic WT recipients, and the recipients were treated with WAY-196025 from the day of transfer. Treatment with WAY-196025 completely blocked EAE development in recipients (Fig. 1A).
AA generated from the cleavage of membrane phospholipids by cPLA2α is processed by COX-1/2 into PGs and by 5-LO into LTs. To determine the relative role of these downstream mediators, we adoptively transferred encephalitogenic T cells as described earlier and therapeutically treated the recipient mice with naproxen, a small-molecule inhibitor of COX-1 and COX-2 (40), or zileuton, a small-molecule inhibitor of 5-LO (41). The naproxen-treated recipient mice had a statistically significant reduction in severity compared with that of the vehicle-treated recipients based on both AUC and MMS (Table I). In contrast, although the onset of EAE was delayed in zileuton-treated recipient mice by 2–3 d, the cumulative disease severity (as measured by AUC) or MMS was not reduced significantly compared with that of the vehicle-treated recipients (Table I). Furthermore, combining naproxen and zileuton therapy to block both the COX-1/2 and 5-LO pathways resulted in the same outcome of clinical disease as treatment with naproxen alone, indicating a potential mutual feedback control of these two pathways with predominance of the COX-1/2 pathway mediators. These results indicate that individual blockade of the COX-1/2 and the 5-LO pathway during the effector phase of EAE is not sufficient to protect mice and that blockade of the upstream enzyme, cPLA2α, is essential for complete protection from EAE.
Treatment . | Mice per Group . | Incidence (%) . | Mean ± SD of Maximum Score . | p Value (Compared with Vehicle for Maximum Score) . | Mean ± SD of AUCa . | p Value (Compared with Vehicle AUC) . | Median Days to Onset . | p Value (Compared with Vehicle for Days to Onset) . |
---|---|---|---|---|---|---|---|---|
Phosal | 6 | 83 | 3.4 ± 0.4 | 37.8 ± 6.4 | 8 | |||
WAY-196025 | 6 | 0 | 0 ± 0 | 5.1e−8 | 0 ± 0 | 1.8e−7 | >24b | 7.5e−8 |
Naproxen | 6 | 59 | 2.5 ± 1.3 | 0.0504 | 20.5 ± 13.1 | 0.0033 | 7 | 0.4227 |
Zileuton | 6 | 65 | 2.9 ± 0.6 | 0.2729 | 24 ± 6.8 | 0.0156 | 10 | 0.3373 |
Naproxen plus zileuton | 6 | 55 | 2.5 ± 0.9 | 0.0504 | 18.8 ± 12.8 | 0.0015 | 8.5 | 0.687 |
Treatment . | Mice per Group . | Incidence (%) . | Mean ± SD of Maximum Score . | p Value (Compared with Vehicle for Maximum Score) . | Mean ± SD of AUCa . | p Value (Compared with Vehicle AUC) . | Median Days to Onset . | p Value (Compared with Vehicle for Days to Onset) . |
---|---|---|---|---|---|---|---|---|
Phosal | 6 | 83 | 3.4 ± 0.4 | 37.8 ± 6.4 | 8 | |||
WAY-196025 | 6 | 0 | 0 ± 0 | 5.1e−8 | 0 ± 0 | 1.8e−7 | >24b | 7.5e−8 |
Naproxen | 6 | 59 | 2.5 ± 1.3 | 0.0504 | 20.5 ± 13.1 | 0.0033 | 7 | 0.4227 |
Zileuton | 6 | 65 | 2.9 ± 0.6 | 0.2729 | 24 ± 6.8 | 0.0156 | 10 | 0.3373 |
Naproxen plus zileuton | 6 | 55 | 2.5 ± 0.9 | 0.0504 | 18.8 ± 12.8 | 0.0015 | 8.5 | 0.687 |
Data representative of two independent experiments.
AUC = sum of all the scores over the course of EAE.
Not estimable because of termination of the experiment, consequently the significance of the test (comparison with the vehicle) is underestimated.
Histological evidence supports the clinical disease outcomes in the vehicle-treated and WAY-196025–treated groups. Microscopic examination of the brains and spinal cords showed remarkable differences between the vehicle-treated and WAY-196025–treated mice (Fig. 1B–E, Table II). The vehicle-treated mice had numerous multifocal to coalescing inflammatory cell infiltrates in the brain and spinal cord (Fig. 1B). Infiltrates were present in the leptomeninges, around blood vessels in the leptomeninges and white matter, and in the parenchyma of the white matter; in the brain they were also localized around the ventricles. Infiltrates in most animals consisted of mononuclear cells, primarily lymphocytes, macrophages, and glial cells. Associated with the inflammatory cell infiltrates in white matter were pallor and vacuolation, consistent with edema and demyelination, and dilated axons were sometimes observed. In contrast, the spinal cords from WAY-196025–treated mice had minimal or no infiltrates (Fig. 1C, Table II). When present, the pathological changes in the WAY-196025–treated mice were much less severe compared with those of the vehicle-treated mice. Luxol fast blue staining, which stains myelin, revealed demyelination in spinal cords from vehicle-treated mice (Fig. 1D, Table II) but no demyelination in spinal cords from WAY-196025–treated mice (Fig. 1E, Table II). Collectively, these data show that cPLA2α blockade during the effector phase of EAE is more efficacious than COX-1/2 or 5-LO blockade alone or in combination.
. | Mean Number of Foci ± SDa (Mean of Affected Animals Only) . | |
---|---|---|
Portion of CNS . | Vehicle . | WAY-196025 . |
Brain | 85.50 ± 20.22 | 107.60 ± 13.88 |
Cervical spinal cord | 36.88 ± 3.11 | 1.60 ± 0.97 |
Thoracic spinal cord | 29.81 ± 0.7b | 1.90 ± 0.65 (1.5)b |
Lumbar spinal cord | 23.79 ± 2.24 | 3.48 ± 1.12 (2.0) |
Mean Demyelination ± SDc | ||
Cervical spinal cord | 1.25 ± 0.25 | 0.0 ± 0.0 (0.0) |
Thoracic spinal cord | 2.00 ± 0.00b | 0.0 ± 0.0 (0.0)b |
Lumbar spinal cord | 3.50 ± 0.29 | 0.0 ± 0.0 (0.0) |
Number Affected/Number Examined | ||
Brain | 4/4 | 5/5 |
Cervical spinal cord | 4/4 | 3/5 |
Thoracic spinal cord | 4/4b | 5/5b |
Lumbar spinal cord | 4/4 | 5/5 |
. | Mean Number of Foci ± SDa (Mean of Affected Animals Only) . | |
---|---|---|
Portion of CNS . | Vehicle . | WAY-196025 . |
Brain | 85.50 ± 20.22 | 107.60 ± 13.88 |
Cervical spinal cord | 36.88 ± 3.11 | 1.60 ± 0.97 |
Thoracic spinal cord | 29.81 ± 0.7b | 1.90 ± 0.65 (1.5)b |
Lumbar spinal cord | 23.79 ± 2.24 | 3.48 ± 1.12 (2.0) |
Mean Demyelination ± SDc | ||
Cervical spinal cord | 1.25 ± 0.25 | 0.0 ± 0.0 (0.0) |
Thoracic spinal cord | 2.00 ± 0.00b | 0.0 ± 0.0 (0.0)b |
Lumbar spinal cord | 3.50 ± 0.29 | 0.0 ± 0.0 (0.0) |
Number Affected/Number Examined | ||
Brain | 4/4 | 5/5 |
Cervical spinal cord | 4/4 | 3/5 |
Thoracic spinal cord | 4/4b | 5/5b |
Lumbar spinal cord | 4/4 | 5/5 |
Foci of 20 or more inflammatory cells were counted.
Representative section of spinal cord of one of these animals from each group is shown in Fig. 1.
Demyelination was scored on the scale 0–4.
Therapeutic dosing with a cPLA2α inhibitor ameliorates Th1- and Th17-mediated EAE in the adoptive transfer model in SJL recipients
It is well established that Th1 cells play a role in EAE (42, 43). Additionally, more recently IL-23 and Th17 cells have emerged as key mediators of EAE (17, 43, 44). We previously showed that both cPLA2α−/− mice (24) and mice treated with WAY-196025 (32) have reduced capacity to mount Th1 and Th17 responses. Although cPLA2α plays a critical role in generation of both Th1 and Th17 cells, it is not clear if it plays a role during the effector phase of the disease when already differentiated encephalitogenic Th1 and Th17 cells are present. To test if therapeutic treatment with WAY-196025 can abrogate disease development in recipient mice adoptively transferred with Th1- or Th17-skewed encephalitogenic cells, we cultured splenocytes from PLP-immunized SJL donors in the presence of either PLP plus IL-12 (Th1) or PLP plus IL-23 (Th17) and adoptively transferred them into naive SJL recipients. The recipient mice were treated with WAY-196025 or vehicle from the day of adoptive transfer. As in the case of adoptive transfer of cells cultured with PLP-only earlier, WAY-196025 treatment significantly reduced disease severity in mice adoptively transferred with either Th1 or Th17 cells, whereas the vehicle treatment in each of these two groups resulted in development of fulminant disease (Fig. 1F). These findings show that therapeutic blockade with cPLA2α inhibitor can ameliorate disease development by both Th1- and Th17-skewed cells.
Therapeutic dosing with a cPLA2α inhibitor is also efficacious in adoptive-transfer EAE in B6 mice
The existing animal models of EAE that are used during drug development have been inconsistent in their ability to predict clinical outcomes of these drugs in MS (6, 32, 45). A case in point is that although the anti-p40 Abs were shown to be efficacious in several EAE models (9–11, 46), they met with failure in the clinical trials of MS (14–16). To address this challenge, we recently reported a novel adoptive transfer model of EAE in B6 mice, in which we demonstrated that anti-p40 Abs were not efficacious during the tissue-damage effector phase of the disease (17), perhaps indicating that this is a more stringent model of EAE. In this model, cells isolated from MOG-immunized donor mice are restimulated with Ag to differentiate and expand the Ag-specific encephalitogenic effector T cells. Importantly, the cells are cultured with Ag and IL-12 under IFN-γ–neutralizing conditions and thereby develop into strongly encephalitogenic effector/memory CD4+ T cells. Adoptive transfer of these T cells into naive syngeneic B6 recipients reproducibly produces fulminant EAE with >90% incidence, thus overcoming the relative resistance of B6 strain to adoptive transfer of EAE (17).
We tested the efficacy of the cPLA2α inhibitors in this model. Recipient mice were treated using therapeutic dosing with WAY-196025 from the day of adoptive transfer of encephalitogenic cells. As in the SJL adoptive transfer model described earlier, WAY-196025 completely blocked disease in B6 recipients, whereas the vehicle-treated group developed fulminant EAE (Fig. 2). These results further demonstrate the efficacy of cPLA2α blockade during the effector phase of EAE in a second mouse strain and when strongly encephalitogenic cells are adoptively transferred.
cPLA2α inhibition in recipient mice does not alter the T cell phenotype but significantly reduces the number of donor-specific encephalitogenic T cells infiltrating the CNS at peak EAE
To investigate the in vivo mechanism of inhibition of cPLA2α during the effector phase of EAE, we used MOG-immunized B6. PL (Thy1.1+) WT mice as donors and naive syngeneic B6 (Thy1.2+) WT mice as recipients in an adoptive-transfer EAE model. Encephalitogenic Thy1.1+/CD4+ T cells, grown with Ag and IL-12 under IFN-γ–neutralizing culture conditions as described earlier, were adoptively transferred into WT B6 recipients (Thy1.2+), and, from the day of adoptive transfer of cells, the recipient mice were treated with WAY-196025. Recipient mice treated with the cPLA2α inhibitor showed no clinical signs of EAE, whereas the recipient mice in the vehicle-treated group developed fulminant disease (data not shown). When recipient mice in the vehicle-treated group reached their peak of disease (average clinical score >2), we harvested the CNS from both groups and analyzed the CNS mononuclear cells by flow cytometry. Donor-specific encephalitogenic Thy1.1+/CD4+ T cells had infiltrated the CNS of Thy1.2+ B6 recipients in both the vehicle-treated and the cPLA2α inhibitor-treated groups (Fig. 3). However, there were 3- to 4-fold fewer Thy1.1+/CD4+ T cells in the CNS of recipient mice treated with the cPLA2α inhibitor compared with that in controls (Fig. 3A–C). Although there were significant differences in the numbers of infiltrating encephalitogenic T cells in the CNS of the two treatment groups (Fig. 3C), the phenotype of infiltrating cells was similar. The vast majority of donor-specific T cells in the CNS of both groups displayed an effector/memory phenotype (Thy1.1+/CD4+/CD45RBlo/CD44hi) (Fig. 3A, 3B). Furthermore, intracellular cytokine staining of cells recovered from the CNS of both treatment groups revealed that a majority of these donor-specific Thy1.1+/CD4+ T cells were IFN-γ+ and only a small percentage (∼1–2%) were IL-17+ (Fig. 3A, 3B). As noted in previous work (17), we found a small but consistent IFN-γ+/IL-17+ double-positive population (∼3–8%) in the CNS of each of the treatment groups (Fig. 3A, 3B).
cPLA2α inhibition reduces microglial activation and improves oligodendrocyte survival in the CNS
Based on the observations that cPLA2α inhibition causes a reduction in number of donor-specific effector T cells recovered from the CNS of WAY-196025–treated recipients at the peak EAE, we hypothesized that cPLA2α inhibition may play a role in the effector T cell responses or APC function or both. We previously reported that MOG-specific T cells from draining lymph nodes (DLN) of MOG-immunized mice treated with WAY-196025 proliferated, upon restimulation with MOG, similarly to vehicle controls (32). This indicates that generation of encephalitogenic Ag-specific T cells from naive precursors is not affected by cPLA2α inhibition. To test whether cPLA2α inhibition has an effect on differentiated effector T cells, we adoptively transferred differentiated effector Thy1.1+/CD4+ T cells to naive B6 Thy1.2+ recipients and treated the recipient mice with vehicle or WAY-196025. Before the onset of EAE, cells from DLN were harvested and analyzed by flow cytometry for the presence of donor-specific Thy1.1+/CD4+ cells. The percentage of Thy1.1+/CD4+ T cells in the peripheral compartment was identical in both WAY-196025–treated and vehicle-treated recipients (Fig. 3D, 3E). Furthermore, when we restimulated cells from these two treatment groups with MOG, they showed identical proliferation (Fig. 3F). These findings indicate that cPLA2α inhibition has little or no effect on proliferation or survival of fully differentiated effector T cells in the peripheral compartment.
Next, we turned our attention to the APC as a potential cell type being affected by cPLA2α inhibition. It was reported that cPLA2α is expressed in the cytosol of APCs, including CNS microglial cells (47, 48). We used the adoptive transfer model in B6 mice and treated recipient mice with either vehicle or WAY-196025 by using a therapeutic dosing regimen as described earlier. At peak disease, we harvested the CNS and performed immunohistochemistry to investigate the activation of microglia and the survival of oligodendrocytes in each of these groups. The CD11b/c equivalent Ab OX-42 recognizes both macrophage and microglial cells in brain. Although OX-42 labels both resting and activated populations of cells, the expression of CD11b/c is profoundly upregulated in activated microglial cells that typically display dendritic cell morphology, and therefore it can be used as a marker of microglial activity.
Treatment with WAY-196025 significantly reduced levels of OX-42staining in the cortex during peak EAE response in B6 mice, suggesting that cPLA2α inhibition may reduce the degree of microglial activation in this brain region during EAE (Fig. 4A, 4B). The sulfatide-recognizing oligodendrocyte marker O4 has been used to define early stages in the maturation of oligodendrocyte progenitors to pro-oligodendrocytes. As shown in Fig. 4C and 4D, WAY-196025 significantly increased the staining for the O4 Ag, suggesting a greater preservation of early oligodendrocytes or the stimulation of oligodendrogenesis. These immunohistochemistry results suggest that cPLA2α inhibitors have some effect on APC activation in the CNS.
cPLA2α inhibition diminishes the ability of APC to support Ag-specific proliferation and cytokine production of effector T cells
CNS Ag-specific T cells undergo a requisite initial reactivation step in the peripheral compartment before trafficking to the CNS (49). To investigate whether cPLA2α inhibition also affected the APC in the peripheral compartment, we immunized WT B10.PL mice with MBP and treated them with either vehicle or WAY-196025 for 10 d, harvested the spleens, and isolated the APCs. Flow cytometry analyses of APCs purified from WAY-196025–treated mice (WAY-196025–APC) showed a substantial increase in the percentage of CD11b+ macrophage population and a substantial decrease in the percentage of CD19+ B cell population compared with those of the APCs purified from vehicle-treated mice (vehicle–APC) (data not shown). This suggested that in addition to blockade of microglia activation in the CNS as demonstrated earlier, cPLA2α inhibition may also result in modulating the APC populations in the peripheral compartment.
It is possible that although the percentage of macrophages in the peripheral compartment was increased in the cPLA2α inhibitor-treated mice, the Ag-presenting function of these cells was affected. To explore potential changes in costimulatory function of these cells, we compared a panel of costimulatory molecules expressed on the APC populations from these two treatment groups (Fig. 5A, 5B and data not shown). APCs from WAY-196025–treated mice had almost six-times higher percentage of PD-L1–expressing macrophages (∼33% CD11b+/PD-L1+) compared with that of vehicle treated mice (∼5% CD11b+/PD-L1+) (Fig. 5A, 5B). To test whether this change in PD-L1 expression had functional consequences, we used Ag-experienced APCs that were isolated from either vehicle-treated (vehicle–APC) or WAY-196025–treated (WAY-196025–APC) mice for coculture with naive CD4+ T cells purified from syngeneic anti-MBP TCR transgenic mice (39, 50) (Tg-T cells). The Tg-T cells showed significantly lower proliferation in response to MBP when cocultured with the Ag-experienced APCs from WAY-196025–treated mice compared with those cocultured with the Ag-experienced APCs from vehicle-treated mice (Fig. 5C). Additionally, these cells produced significantly more IFN-γ than that of the control group (Fig. 5D). We observed no differences in IL-10 or TNF production from these two groups (Fig. 5E, 5F). Therefore, in the peripheral compartment, cPLA2α inhibition does not appear to curtail the expansion of macrophages in treated mice and may even support their increase. However, cPLA2α inhibition has a suppressive effect on the Ag-presentation function of these macrophages, resulting in reduced proliferation and altered differentiation of Ag-specific encephalitogenic effector cells.
cPLA2α inhibition induces changes in APC function that may promote the development of Tregs
Based on our observation that cPLA2α inhibition may cause increased populations of CD11b+/PD-L1+ APCs in the peripheral compartment with an altered T cell differentiating function, we investigated whether these APCs would support the generation of Tregs when cultured in presence of TGF-β . Ag-experienced APC from spleens of either vehicle-treated mice or WAY-196025–treated mice were cocultured with purified Tg-T cells in the presence of Ag plus suppressive levels of TGF-β. We measured proliferation and cytokine production. As expected, for cultures with MBP plus TGF-β, the proliferation of Tg-T cells was substantially curtailed in both groups compared with that of control MBP-only cultures described earlier (compare Figs. 5C and 6A). In contrast to MBP-only cultures above (Fig. 5C), the proliferation of Tg-T cells cultured with MBP plus TGF-β was greater in coculture with WAY-196025–APC compared with that of vehicle–APC (Fig. 6A). Moreover, the Tg-T cells cultured with MBP plus TGF-β in presence of WAY-196025–APC showed increased production of IFN-γ and IL-10 and a moderate increase in TNF compared with the Tg-T cells cultured with MBP plus TGF-β in presence of vehicle–APC (Fig. 6B–D).
Although the relative percentages of Foxp3+ versus Foxp3− expressing T cells were not different between these two groups (compare Fig. 6F and 6G), there was approximately a 2-fold greater percentage of Foxp3+ T cells generated when Tg-T cells were cultured with MBP plus TGF-β in the presence of WAY-196025–APC compared with that in presence of vehicle–APC (Fig. 6F, 6G). These results may suggest that the WAY-196025–APC supports differentiation and proliferation of Ag-specific Tregs (Foxp3+ and IL-10 production) in presence of a suppressive cytokine environment, such as high levels of TGF-β (Fig. 6E).
Therapeutic dosing with cPLA2α inhibitor abrogated relapses in EAE in the SJL direct disease model
Unlike patients with primary progressive MS, patients with relapsing–remitting MS display a characteristic relapse–remission of disease symptoms, and this hallmark feature is also observed in the direct disease model of EAE in SJL mice. Multiple studies have indicated that this relapsing–remitting phenotype is due to epitope spreading (51, 52). Because WAY-196025 showed efficacy during the effector phase of EAE in two different adoptive transfer models described earlier, we wanted to expand these observations and test if WAY-196025 could prevent relapses of EAE in the SJL direct disease model as well. WT SJL mice were immunized with PLP/CFA and subsequently dosed using two different therapeutic dosing regimens, either starting from peak of disease (peak dosing) (Fig. 7A) or before the first relapse (relapse dosing) (Fig. 7B). Prior to the start of dosing, mice were randomly recruited into either vehicle or WAY-196025 treatment groups for each of these two different therapeutic dosing regimens, and scoring was continued on a daily basis. We observed that compared with the mice in the two vehicle treatment groups, the mice in the two WAY-196025 treatment groups showed remarkable and rapid recovery from disease (Fig. 7A, 7B). This recovery was accompanied by almost complete reversal of paralysis and ataxia and by abrogation of subsequent disease relapses for the entire duration of treatment in both WAY-196025 dosing groups (peak dosing and relapse dosing) (Fig. 7A, 7B). This demonstrates that blocking cPLA2α helps to prevent subsequent relapse of disease with a notable recovery from ataxia and paralysis.
Discussion
Understanding the mechanism of cPLA2α in generation of multiple proinflammatory mediators during the various stages of EAE may help in the development of more effective therapies for MS. In the current study, we have used models of EAE (i.e., adoptive transfer models in SJL and B6 mice as well as relapsing–remitting EAE in SJL mice) that are focused on the tissue-damage effector phase of the disease, which is more relevant to the MS pathology (14, 17), to demonstrate that a highly selective small-molecule inhibitor of cPLA2α, WAY-196025, abrogated disease development. We also explored the mechanism of cPLA2α blockade in these models demonstrating that cPLA2α plays a role in APC function such that its blockade reduced Ag-specific activation of T cells, both in the peripheral and in the CNS compartments, resulting in amelioration of disease.
The various metabolites produced by the cPLA2α pathway could be responsible for mediating the different pathological features of MS. The primary function of cPLA2α enzyme is to release AA from membrane lipids, which is then converted primarily to PGs via COX-1/2 and LTs via 5-LO pathways. PGs such as TXA4 promote lymphocyte proliferation, adhesion, and enhance macrophage function (53, 54). Additionally, it has been shown that PGE2 may regulate Th17 cell differentiation and function in both humans and mice (34). LTs such as LTB4 can recruit macrophages to the CNS and activate their phagocytic function, which produce proinflammatory cytokines and NO and cause further damage to the myelin sheath (53, 55). LTB4 is a potent chemoattractant of myeloid cells as well as CD4 and CD8 effector cells (56–58). LTB4 also promotes leukocyte adhesion to endothelial cells and extravasation into tissues (53).
To understand the relative contribution of these enzymes to the pathogenesis of EAE during the tissue-damage effector phase, we used the adoptive transfer model of EAE and selectively blocked COX-1/2 or 5-LO or cPLA2α pathways in recipient mice. Blockade of COX-1/2 (using naproxen) reduced severity of EAE, whereas blockade of 5-LO (using zileuton) delayed onset and reduced cumulative EAE severity but did not reduce maximum severity of disease. Strikingly, in the same experiment, blockade of cPLA2α (using WAY-196025) resulted in complete protection from the adoptively transferred EAE. In fact, combined therapy of naproxen and zileuton was less effective than WAY-196025 in ameliorating the preclinical disease in recipient mice. Thus, inhibition of cPLA2α provides better clinical outcome compared with inhibition of the two downstream pathways (COX-1/2 or 5-LO or both) during the tissue-damage effector phase of EAE.
One possible explanation for the greater efficacy of cPLA2α inhibition, compared with blockade of COX-1 and -2 or 5-LO, is that selective blockade of individual downstream pathways could result in shunting of the excess AA into the alternate unblocked pathway, thereby increasing the proinflammatory mediators produced by that pathway and contributing to disease pathology. Furthermore, in addition to COX-1/2 and 5-LO, a third pathway operates downstream of cPLA2α and results in generation of LPC and PAF. The proinflammatory mediators of this pathway could provide another possible explanation for the lack of efficacy of blockade of COX-1/2 and 5-LO in EAE. LPC, which is generated by action of cPLA2α, not only acts as a myelinolysis agent, but also, in late stages of demyelination, damages the exposed axonal membranes (55, 59). LPC is also a chemoattractant for effector T cells and induces expression of the cell adhesion molecules ICAM-1 and VCAM-1 by CNS endothelial cells and rapidly opens the blood–brain barrier (59). Downstream of LPC is PAF, which plays an important role in EAE, in that PAF receptor-deficient mice have reduced incidence and severity of EAE (25). More recently it has been demonstrated that PAF accumulates in the spinal cords of EAE mice, and this is likely due to increased activity of the cPLA2/acetyl-CoA:lyso-PAF acetyltransferase axis present in infiltrating macrophages and activated microglia (60).
We have previously shown that cPLA2α−/− B6 mice are resistant to EAE induction (24). Furthermore, we have demonstrated that prophylactic treatment during the induction phase of EAE with WAY-196025 prevented disease development in adult WT mice (32), thereby confirming that resistance to EAE induction in cPLA2α−/− mice was not a result of absence of cPLA2α during embryonic, fetal, or postnatal development. In this study, we used the adoptive transfer and relapsing–remitting models of EAE to show that WAY-196025 is efficacious by therapeutic dosing, confirming that cPLA2α may be an important therapeutic target for MS.
To understand the mechanism of action of cPLA2α during the course of EAE pathology, we investigated the effect of cPLA2α blockade on various cells involved in EAE pathogenesis. In our adoptive transfer model, a significantly reduced percentage of encephalitogenic donor-specific Thy1.1+/CD4+ cells were recovered from CNS of recipients treated with cPLA2α inhibitor, corresponding with the absence of disease in these mice. This suggests that cPLA2α inhibition plays a role in the effector T cell responses and/or APC function. We previously reported that MOG-specific T cells from DLN of MOG-immunized mice treated with WAY-196025 proliferated, upon restimulation with MOG, similar to vehicle controls (32). This indicated that generation of encephalitogenic Ag-specific T cells from naive precursors was not affected by cPLA2α inhibition. To test whether cPLA2α inhibition has an effect on differentiated effector T cells, we adoptively transferred Thy1.1+/CD4+ T cells to Thy1.2+ recipients and treated the recipient mice with vehicle or WAY-196025. Before the onset of EAE, cells from DLN were harvested and analyzed by FACS for the presence of donor-specific Thy1.1+/CD4+ cells. The percentage of Thy1.1+/CD4+ T cells in the peripheral compartment was identical in both WAY-196025–treated and vehicle-treated recipients. Furthermore, when we restimulated cells from these two treatment groups with the appropriate CNS Ag, they showed identical proliferation. These findings indicate that cPLA2α inhibition has little or no effect on proliferation or survival of differentiated effector T cells in the peripheral compartment. However, we cannot fully rule out a potential long-term effect of cPLA2α inhibition on T effector cell survival or trafficking.
We next explored the effect of cPLA2α on APC function. At peak EAE, when compared with vehicle-treated recipient mice, the CNS of WAY-196025–treated recipient mice showed improved oligodendrocyte survival and attenuated microglial activation. This suggests that cPLA2α acts on multiple cell types and plays multiple roles in the CNS during the course of the effector phase of EAE, each of which could contribute to the pathogenesis of the disease. Further investigation of a potential role of cPLA2α in APC function demonstrated that APC fractions purified from WAY-196025–treated donor mice (WAY-196025–APC), compared with those purified from vehicle-treated donors (vehicle–APC), show increased percentage of CD11b+ macrophages that expressed PD-L1 on their surfaces. Compared with the vehicle–APC, the WAY-196025–APC were much less capable of supporting MBP-specific in vitro proliferation of purified naive responder CD4+ T cells isolated from anti-MBP TCR transgenic mice (Tg-T cells). Additionally, the WAY-196025–APC/Tg-T cell cocultures produced significantly greater amounts of IFN-γ compared with those of vehicle–APC/Tg-T cell cocultures, explaining, at least in part, the reduced proliferation. PD-L1 is a well-characterized negative regulator of T cell proliferation (61). Furthermore, recent studies have shown that PD-L1 expressed on myeloid-derived APCs in the CNS negatively regulates the PD-1+ T cell responses during acute relapsing EAE (62). Therefore, one of the potential mechanisms by which cPLA2α blockade may ameliorate disease could be by modulation of APC function to prevent myelin Ag-specific proliferation of effector T cells.
Additional studies elucidated the extent of modulation of APC function by cPLA2α inhibition. Thus, we observed that APC fractions purified from WAY-196025–treated donor mice (WAY-196025–APC), when cocultured with Tg-T cells in presence of TGF-β, were better at skewing the Tg-T cells into Foxp3+ Tregs compared with that of APC fractions purified from vehicle-treated donor mice (vehicle–APC). This appears to indicate that cPLA2α inhibition affects APC function, and this not only can result in curtailment of effector T cell proliferation, but also, in presence of suppressive levels of TGF-β, can generate increased numbers of Foxp3+ Treg cells. A recent report showed that PD-L1 on APCs has an obligatory role in controlling induced-Treg cell development and function in vivo (63). We are continuing to explore this potential mechanism of cPLA2α inhibition.
Both Th1 and Th17 cells have been implicated in the pathogenesis of EAE and MS (64). We have shown that cPLA2α−/− mice have reduced capacity to generate Th1 cells (24), and cPLA2α inhibition in direct EAE models diminishes the production of Th1-type and IL-17 cytokines (32). We show in this study that therapeutic treatment with WAY-196025 ameliorated the disease in SJL mice adoptively transferred with PLP-specific effector cells that were skewed under either Th1 or Th17 conditions in vitro. Additionally, we demonstrate the ameliorating effect of cPLA2α inhibition in a relapsing–remitting model of EAE in SJL mice. In this model, therapeutic dosing with WAY-196025, dosed either from peak EAE or after the primary episode, rapidly reduced the disease severity and protected mice from subsequent relapses compared with vehicle controls.
Taken together, our data suggest that during the tissue-damage effector phase of disease, cPLA2α blockade is the better therapy for rapid amelioration of EAE in mice. Moreover, the mechanism of action of cPLA2α suggests that it plays a critical role in APC activation and function resulting in targeted reduction of encephalitogenic effector T cell numbers and/or function. Therefore, cPLA2α inhibition in APCs could be beneficial during ongoing MS, which involves continuous restimulation of encephalitogenic effector T cells in the CNS as well as in the DLN in the peripheral compartment. It is intriguing to note that this modulation of APC function by inhibiting the cPLA2α pathway predisposes the APC to generate Foxp3+ Tregs under a suppressive cytokine milieu. We are continuing to investigate this and other potential mechanisms of cPLA2α inhibition.
Acknowledgements
We thank Dr. Susumu Tonegawa for anti-MBP TCR transgenic mice. We thank Dr. Kyriaki Dunussi-Joannopoulos and Dr. Takao Shimizu for useful discussions and suggestions. We also thank Stephen Benoit, Jeffrey Pelker, and Ming Zhu for technical contributions. Statistics on the microscopic differences were prepared by Dr. Youping Huang, and statistics for Table II were prepared by Dr. Shuguang Huang.
Footnotes
Abbreviations used in this article:
- AA
arachidonic acid
- AUC
area under the curve
- B6
C57BL/6
- COX
cyclooxygenase
- cPLA2α
cytosolic phospholipase A2α
- DLN
draining lymph nodes
- EAE
experimental autoimmune encephalomyelitis
- 5-LO
5-lipoxygenase
- LPC
lysophosphatidylcholine
- LT
leukotriene
- LTB4
leukotriene B4
- MBP
myelin basic protein
- MMS
mean maximum severity
- MOG
myelin oligodendrocyte glycoprotein
- MS
multiple sclerosis
- PAF
platelet-activating factor
- PLP
proteolipid peptide
- Treg
regulatory T cell
- WT
wild-type.
References
Disclosures
The authors have no financial conflicts of interest.