Interleukin (IL)-4 indirectly suppresses IL-2 production by human T lymphocytes via peroxisome proliferator-activated receptor gamma activated by macrophage-derived 12/15-lipoxygenase ligands.

The respective development of either T helper type 1 (Th1) or Th2 cells is believed to be mediated by the effects of cytokines acting directly on Th precursors (Thp). We have generated evidence for an indirect monocyte-dependent immunoregulatory pathway. Recently, interleukin (IL) 4 has been shown to produce "new" potential peroxisome proliferator-activated receptor gamma (PPARgamma) ligands by inducing macrophage 12/15-lipoxygenase (12/15-LO). We have shown previously that the activated PPARgamma is a profound inhibitor of IL-2 transcription in human T lymphocytes. It is hypothetically possible that IL-4 might indirectly affect IL-2 production by Thp cells via macrophage-derived PPARgamma ligands. Using human monocytes and T lymphocytes from same donors, we have found that monocyte 12/15-LO products mediate the indirect inhibitory effect of IL-4 on anti-CD3- or phytohemagglutinin/phorbol 12-myristate 13-acetate-stimulated IL-2 production by T lymphocytes. We further analyzed which major 12/15-LO metabolites contributed to the above inhibition. 13-Hydroxyoctadecadienoic acid (13-HODE), a 12/15-LO product, markedly blocked IL-2 production by human blood T lymphocytes, but not Jurkat T cells. Moreover, the IL-4-conditioned macrophage medium contained a sufficient amount of 13-HODE and anti-13-HODE antibody indeed neutralized the inhibitory effects of the IL-4-conditional medium on T-cell IL-2 production. Using human T lymphocytes and the PPARgamma-transfected Jurkat T cells, we demonstrated the specific inhibition by 13-HODE of the transcription factors NFAT (nuclear factor of activated T cells) and nuclear factor kappaB, the IL-2 promoter reporter, and IL-2 production. However, 15-hydroxytetraenoic acid had little inhibitory effect. The potency of such inhibitory effects correlates well with the capability of the above metabolic lipids to activate PPARgamma. These data provide a mechanism whereby IL-4 may indirectly affect Thp function via PPARgamma activated by macrophage products of the 12/15-LO pathway.

T helper (Th) 1 lymphocytes can be divided into two functional subsets consisting of Th1 and Th2 cells on the basis of the immunoregulatory cytokines that these T cells produce (1)(2)(3). Some of these immunoregulatory cytokines possess cross-regulatory properties and can enhance or suppress cytokine production by Th1 or Th2 subset. Thp cells are the pluripotent precursors of Th1 and Th2 cells (4). Moreover, the development of either Th1 or Th2 helper cells is believed to be determined by the effects of cytokines directly on helper Thp cells. IL-4 is principally produced by helper T cells of the Th2 phenotype. IL-4 has been shown to be a pleiotropic lymphokine with an array of biologic effects on multiple cell lineages (5,6). IL-4 can function as a growth factor for activated T cells including promoting T cell proliferation and IL-2 production (7,8). Importantly, all of the effects of IL-4 on human T cells have been inferred from experiments using mixed population of cells. Inasmuch as IL-4 has been shown to have effects on a variety of cell types, including monocyte/macrophages and B cells that can function as accessory cells. IL-4 can inhibit IL-2 synthesis by concanavalin A-stimulating CD4ϩ human T cells in the presence of accessory cells (9,10). It is hypothetically possible that the effect of IL-4 on human T cell activation is indirect and mediated by one of these accessory cells.
The monocyte/macrophage is well recognized as essential in the regulation of lymphocyte function. Some aspects of this regulation involve the release of soluble mediators by monocyte macrophages. Interestingly, IL-4 has been shown to induce 12/ 15-lipoxygenase (LO) in monocytes/macrophages, which in turn produce "new" potential peroxisome proliferator-activated receptor ␥ (PPAR␥) ligands (11). PPAR␥ is a unique member of liganddependent nuclear receptor family that has been implicated in the modulation of critical aspects of development and homeostasis, including adipocyte differentiation, glucose metabolism, and macrophage development and function (12)(13)(14)(15). Previously, we have shown the expression of PPAR␥ in human T cells. Using a range of synthetic and natural PPAR␥ ligands, including troglitazone and 15d-prostaglandin J 2 , we have demonstrated that activation of PPAR␥ could block IL-2 production in T cells by inhibiting NFAT-mediated transcription of the IL-2 gene. Activated PPAR␥ physically associated with NFAT, blocking IL-2 promoter activity. This represented a novel mechanism and function of the PPAR␥ nuclear receptor in T cell biology (16). It is well known that IL-4 exerts immunomodulatory effects on monocytes and T cells. These observations led us to hypothesize that IL-4 might indirectly affect the production of IL-2 by Thp helper cells by inducing the production of these potential PPAR␥ ligands by macrophage 12/15-lipoxygenase, which in turn interfere with the subsequent development of T helper cells.

EXPERIMENTAL PROCEDURES
Materials-Human IL-4 was from PeproTech Inc. 13-HODE and 15-HETE were from Cayman Chemical. PD146176 and troglitazone were the gifts from Dr. J. Cornicelli and M. A. Caballero of Parke-Davis. Anti-13-HODE antibody and 13-HODE immunoassay kit were from Oxford Biomedical Research.
Cell Culture-Human peripheral blood monocytes and T lymphocytes were obtained from same healthy donors and cultured in RPMI with 10% fetal calf serum (Sigma), 2 mM L-glutamine, and penicillinstreptomycin (50 IU/ml and 50 g/ml, respectively; Invitrogen). Jurkat T cells were maintained under the same conditions. 3T3-L1 preadipocytes were cultured, maintained, and differentiated as described previously by Hamm et al. (17) and Shao and Lazar (18).
IL-2 Measured by ELISA-T cells were grown to ϳ2.5 ϫ 10 6 cells/ml and treated with anti-CD3 or PHA/PMA in the presence or absence of the different ligands for 24 h. Cell supernatants were collected and assayed for IL-2 by ELISA using Endogen kits (Wolburn).
Determination of 13-HODE Levels by ELISA (19,20)-13-HODE was extracted from each sample at 4°C as follows. The solution was acidified to a pH of 3.5-4.0. The organic phase of the solution was extracted using water-saturated ethyl acetate. Samples were dried completely under nitrogen, then reconstituted with a mixture of 25 l of methanol, 975 l of dilution buffer, and 50 l of chloroform. The pH was adjusted to 7.2, and the samples were stored at Ϫ20°C. The plates pre-coated with anti-13-HODE antibodies were used to measure 13-HODE levels at room temperature. Serial dilutions of sample extracts were prepared, and 100-l volumes of each dilution were added to wells. An aliquot of 100 l of a 13-HODE-horseradish peroxidase conjugate (1:1000) was added to each well, and the plates were incubated for 2 h at room temperature. Wells were washed twice with wash buffer, and 200 l of 3,3Ј,5,5Ј-tetramethylbenzidine reagent was added. After incubated for 20 min, the reaction was terminated by adding 50 l of 1 N sulfuric acid. The absorbance was measured using a microtiter plate reader at 450 nm.
Transient Transfection-Transient transfections of human blood T lymphocytes upon stimulation with a concentration of PHA (1 g/ml) were performed by the method described by Hughes et al. (23) and Cron et al. (24). For Jurkat T cells, transfection was performed according to the manufacturer's instructions for FuGENE 6 (Roche Molecular Biochemicals). Briefly, FuGENE 6 was mixed with the plasmid DNA at the ratio of 2:1. The mixture was incubated for 20 min at room temperature and then added to cell culture flask containing 2 ϫ 10 7 cells. After 6 h, cells were washed twice with RPMI 1640, replaced in normal medium, and seeded in a 12-well plate. Cells were treated with or without different ligands for an additional 24 h in the present or absence of PHA/PMA.
Luciferase Reporter Assays (25)-The transfected cells were pelleted, lysed, and then centrifuged at 12,000 ϫ g in a microcentrifuge for 2 min at 4°C. The supernatant was transferred into a new tube, and 20 l of lysate was mixed with 100 l of luciferase assay reagent in cuvettes for the luminometer. The luciferase assay measurement was normalized by the protein amount.

IL-4 Indirectly Inhibits Anti-CD3-or PHA/PMA-stimulated IL-2 Production by T Lymphocytes via Monocyte 12/15-LO
Products-To examine the possibility and physiological relevance of the regulation of the soluble mediators released by IL-4-treated monocyte/macrophages on T lymphocyte activation, we first tested the effect of products from IL-4-treated macrophages on IL-2 production by fresh human T lymphocytes. Human peripheral blood monocytes and T lymphocytes were obtained from the same donor. Human blood monocytes were cultured with or without IL-4 for 96 h (11). Human T lymphocytes were stimulated with the human anti-CD3 antibody or PHA/PMA plus the above macrophage-conditioned medium. After 24 h, the supernatants were collected and tested their IL-2 content by ELISA. As shown in Fig. 1a, T cells stimulated with anti-CD3 or PHA/PMA in conditioned medium from IL-4-treated macrophages produced significantly less (Ϫ62.2% or Ϫ44.5%, respectively) IL-2. However, this inhibition was reversed by medium conditioned by macrophages treated with IL-4 and PD146176, the specific 12/15-LO inhibitor, when compared with non-IL-4-treated macrophages. Direct treatment with PD146176 or IL-4 on purified T cells had no observable inhibitory effects (data not shown). Furthermore, Western blot analysis showed the expression of 12/15-LO by IL-4-induced monocytes, but not by blood primary T lymphocytes or Jurkat T cells (Fig. 1b), which was consistent with the previous reports that IL-4 induces 12/15-LO expression on human monocyte (27), but not human lymphocytes (28). T-lymphocytes isolated from human blood probably do not metabolize polyunsaturated fatty acid via the lipoxygenase pathway (29). These findings suggest that monocyte 12/15-LO products con-FIG. 1. a, IL-4-treated macrophage products modulate IL-2 production by fresh human T lymphocytes. Human monocytes were cultured with or without IL-4 and treated with PD146176, the specific 12/15-LO inhibitor, for 96 h. Human T lymphocytes were cultured with the above macrophage-conditioned medium in the absence or presence of anti-13-HODE, and stimulated with anti-CD3 or PHA/PMA. After 24 h, the supernatants were collected and tested for IL-2 titer by ELISA. b, comparison of 15-LO protein expression in human peripheral blood monocytes, peripheral blood T lymphocytes, and Jurkat T cells.
tribute to an indirect inhibitory effect of IL-4 on IL-2 production by T lymphocytes.
Effect of 12/15-LO Metabolites on IL-2 Production by Fresh Human T Lymphocytes-12/15-Lipoxygenase generates bioactive lipid mediators from free polyunsaturated fatty acids in human monocytes/macrophages (30). 13-HODE and 15-HETE are the major metabolites formed from exogenous linoleic acid and arachidonic acid. To clarify the mechanism underlying these inhibitory effects of macrophage 12/15-lipoxygenase products on T cell activation, we compared the direct effects of the above IL-4/macrophage-induced 12/15-lipoxygenase products on T cell activation. Human peripheral blood T cells were stimulated with PHA/PMA and cultured with various concentrations of different ligands for 24 h. As shown in Fig. 2, 13-HODE markedly decreased anti-CD3-or PHA/PMA-induced IL-2 synthesis in a dose-dependent manner. In contrast, 15-HETE showed very weak inhibitory effects on anti-CD3-or PHA/PMA-induced T cell activation. These results suggest 13-HODE, but not 15-HETE, is a major bioactive mediator, present in 12/15-lipoxygenase macrophage products interfering with T lymphocyte activation.
A reasonably high level of exogenous 13-HODE is required to achieve significant inhibition of IL-2 production by T cells. Thus, it is critical to determine whether the conditioned medium does contain a sufficient amount of 13-HODE. We measured the concentration of 13-HODE in the conditioned medium by a competitive ELISA. As shown in Table I, IL-4 could increase the amount of 13-HODE to an approximately concentration of 40 M, which was corresponding well with ED 50 of exogenous 13-HODE used in our experiments. However, in the presence of PD146176, the formation of 13-HODE induced by IL-4 was significantly decreased. Furthermore, because the anti-13-HODE antibody used in this study was reported previously to react with 13-HODE in human prostate tissues, we thus asked if the anti-13-HODE antibody does affect the inhibition of the IL-4-induced monocyte conditional medium on IL-2 production by T-cells. Fig. 1a also showed the addition of anti-13-HODE antibody indeed neutralized the inhibitory effects of the IL-4-induced monocyte conditional medium on IL-2 production by anti-CD3-or PHA/PMA-stimulated T-cells. By contrast, when anti-13-HODE antibody was replaced by control normal goat serum, the decrease in T cell IL-2 production caused by IL-4-induced monocyte conditional medium remained at the same level (data not shown). These results confirmed 13-HODE produced by IL-4-induced macrophages is significant enough to down-regulate T cell activation.
Inhibition of 12/15-LO Metabolites on IL-2 Production and Promoter Activity in PPAR␥-dependent Manner-Because IL-4 strongly produced novel PPAR␥ ligands by 12/15-LO in monocytes (11), we determined whether inhibition of 12/15-lipoxygenase products on T lymphocytes was through PPAR␥. We performed Western blot analysis to confirm the expression of PPAR␥ on human T lymphocytes. As shown in Fig. 3a, differentiated 3T3-L1 cells, which express both PPAR␥1 and PPAR␥2 isoforms (17,18,31,32), were used as a positive control for the expression of PPAR␥. Human T lymphocytes and monocytes, but not Jurkat T cells, contained PPAR␥ protein, which was consistent with the Northern blot results described by Greene et al. (33). We next tested the effect of 13-HODE on IL-2 production by Jurkat T cells, which lacks PPAR␥, to verify the necessity of PPAR␥ expression for the repressive effects of PPAR␥ ligands observed on human T lymphocytes. Fig. 3b showed 13-HODE and troglitazone did not decrease the production of IL-2 by Jurkat T cells. These results indicated that PPAR␥ might be involved in inhibition of 12/15-LO metabolites, as PPAR␥ ligands, on human T lymphocytes.
To determine whether the inhibitory effect of 12/15-lipoxygenase products on IL-2 synthesis can be ascribed, at least in part, to disruption of IL-2 promoter (34, 35) activity, the purified human blood T lymphocytes, upon stimulation with a concentration of PHA (1 g/ml) that is insufficient to cause significant IL-2 secretion (23, 24), were transfected with IL-2 promoter luciferase reporter constructs. PMA/PHA treatment resulted in a marked increase in IL-2 promoter activity. 13-HODE, but not 15-HETE, was able to largely block IL-2 promoter activity in human T lymphocytes (Fig. 4), which was in parallel to the observation on the inhibition of other PPAR␥ ligands on PPAR␥-transfected Jurkat T cells (16). These data suggest that the inhibitory effect of 12/15-lipoxygenase products on IL-2 synthesis was caused by disruption of IL-2 promoter activity in human T lymphocytes even in the absence of overexpression of PPAR␥.

Effect of 12/15-LO Products on DNA Binding and Transcriptional Activation of NFAT and NF-B-
The IL-2 promoter contains five NFAT binding sites, an NF-B binding site, and two Oct-1 sites (35)(36)(37). It has been shown previously that, among these factors, NFAT is obligatory for the induction of IL-2 expression during T cell activation (38 -40). Previously, we have shown that activation of PPAR␥ with 15d-prostaglandin J 2 and troglitazone block NFAT by forming a complex (16). Therefore, we evaluated the effect of 13-HODE and 15-HETE on DNA binding and transcriptional activity of NFAT. As shown by EMSA in Fig. 5a, the specific binding of an NFAT probe corresponding to the human IL-2 promoter was strongly induced by PHA/PMA in human T lymphocytes, which could be shifted by anti-NFATc. Equivalent nuclear extracts from 13-HODE-treated cells displayed diminished binding capacity by the 32 P-radiolabeled probes. This indicated 13-HODE could block the DNA binding activity of NFAT. In contrast, Western blot analysis showed that the expression of NFAT (Fig. 5b) did not change with 13-HODE and 15-HETE treatment. Furthermore, the transcriptional activation of NFAT was measured on PPAR␥-transfected Jurkat T cells (16, 41) by a reporter con-   tivity of NFAT induced by PHA/PMA in the presence of PPAR␥ overexpression (Fig. 5c). However, 15-HETE did not significantly inhibit DNA binding and transcriptional activity of NFAT. Interestingly, the inhibitory effect of the above 12/ 15-LO metabolic lipids correlates well with their capability to activate PPAR␥.
To determine whether transcription factor NF-B was equally inhibited by 13-HODE and 15-HETE in T cells, the DNA binding and transcriptional activity of NF-B were examined. For this case, nuclear cell extracts were incubated with the NF-B DNA binding element and supershifted with the p65 or p50 antibody to confirm the identity. The antibody directed against p50 (not p65) could significantly supershift the NF-B DNA binding. 13-HODE, but not 15-HETE, was effective at inhibiting PHA/PMAinducible NF-B DNA binding activity (Fig. 6a), although the above 12/15-LO metabolic lipids did not affect the protein level of NF-B (Fig. 6b) in human T lymphocytes. Moreover, we analyzed NF-B transactivation by a luciferase reporter gene. As shown in Fig. 6c, NF-B transcription activity following PHA/PMA stimulation was inhibited by 13-HODE but not 15-HETE in the overexpression of PPAR␥. Thus, there appears to be a selective disruption of the transcriptional regulation of the IL-2 promoter mediated by specific 12/15-lipoxygenase products repressing IL-2 production by human T cells. DISCUSSION IL-2 is primarily a product of the Thp and Th1 subclasses of helper T cells. IL-2 production is indicative of T cell activation and is the major autocrine and paracrine growth factor for T cells. Therefore, the regulation of IL-2 production is a key event for control of T cell survival, clonal expansion, and functional differentiation and development (34,35). It has been reported that the Th2 cytokine IL-4 plays a critical role in the development of T helper cells by regulating IL-2 production by Thp cells in both a direct and an indirect manner (4 -7). Moreover, IL-4 largely potently decreases the transcriptional activation of IL-2 in response to concanavalin A in normal human peripheral blood T cells in the presence of 10% accessory cells. These observations suggest that monocytes/macrophages, as typical accessory cells, are of central importance in the initiation, development, and outcome of the immune response and are also a target for type 1 and type 2 cytokines in the immune response (9 -10). The data presented in this study support an important role for macrophages in the indirect pathway of IL-4 in inhibiting IL-2 production by fresh human peripheral blood T cells. Furthermore, we provided evidence that monocyte/ macrophage 12/15-lipoxygenase products mediate this indirect inhibitory effect of IL-4 on IL-2 production by T lymphocytes and requires the expression of PPAR␥ in Thp lymphocytes.
IL-4 is a potent modulator of monocyte function through modulating the metabolism of polyunsaturated fatty acids. IL-4 can induce 12/15-lipoxygenase (42) and suppress prostaglandin H synthase (cyclooxygenases)-2 (43,44), but phospholipase A 2 is not coupled to IL-4 receptor signaling (45) in monocytes. Very recently, Spanbroek et al. (46) reported that IL-4 determines eicosanoid formation in differentiating dendritic cells derived from hematopoietic progenitor cells and human blood monocytes by up-regulation of 15-LO and down-regulation of 5-LO. The enzyme 15-lipoxygenase is unique among the human lipoxygenases in that it is capable of oxygenating polyenoic fatty acids esterified to membrane lipids or lipoproteins, and hence it may have biological roles distinct from its action on free arachidonic acid. 12/15-Lipoxygenase has been implicated in a number of cellular processes, including degradation of intracellular organelles and oxidation of low density lipoprotein, and in a wide variety of disease states such as atherosclerosis, asthma, and psoriasis (27). 15-LO was also shown to mediate nonsteroidal anti-inflammatory drug-induced apoptosis independently of cyclooxygenase-2 in colon cancer cells (47). Human 15-lipoxygenase is a potential effector molecule for IL-4. Although the enzyme and incubated with medium (Ϫ) or PHA/PMA (ϩ) for 2 h at 37°C. Nuclear extracts corresponding to 5 g of protein were incubated with a 32 P-labeled oligonucleotide NF-B probe. Arrow indicates migrational location of each DNA complex. Supershift analyses were performed by the addition of 2 l of IgG against NF-B p65 or p50. b, the above nuclear extracts from human T lymphocytes were separated by SDS-PAGE and immunoblotted by anti-NF-B p65 (upper) or p50 (lower). c, Jurkat cells were co-transfected with an NF-B luciferase reporter construct and a PPAR␥ expression plasmid. Cells were treated with above ligands, stimulated by PHA/PMA, as shown, and collected for analysis of reporter gene activity 24 h later. activity of 12/15-LO is low or undetectable in quiescent peripheral blood monocytes, IL-4 specifically induces 12/15-LO mRNA, protein expression (Fig. 1b) and enzymatic activity and dramatically increased the formation of 13-HODE and 15-HETE in cultured monocytes probably through a Stat6-dependent pathway (42). 13-HODE may also be present in even higher amounts because linoleic acid may be the preferred substrate for human 15-LO. Using a competitive ELISA, we have found IL-4 indeed increased the level of 13-HODE in monocytes (Table I). Moreover, anti-HODE also could neutralize the inhibitory effect of IL-4treated monocyte conditional medium on IL-2 production by T cells (Fig. 1a). Recently, Nagy et al. (48) and Tontonoz et al. (49) reported that 13-HODE, which is formed by 15-LO and by oxidation of lipid component in cells, is a potent endogenous activator and ligand for PPAR␥. However, 15-HETE was only a weak activation of PPAR␥. Using human T lymphocytes and PPAR␥transfected Jurkat T cells, we confirmed the capability of the above 12/15-LO products to activate PPAR␥ in human T lymphocytes.
PPAR␥ is a ligand-dependent transcription factor and activated by diverse synthetic and naturally occurring substances. Although most studies concern the regulation of glucose and lipid metabolism (48 -50) by PPAR␥, research studies over the past year have suggested that this nuclear receptor might also play a number of additional roles in inflammation, atherosclerosis, and cancer (51)(52)(53)(54)(55). Previously, we have reported the role of PPAR␥ in T lymphocyte activation including inhibition of IL-2 production and PHA-induced cell proliferation. In this study, we have been able to confirm the expression of PPAR␥ in human blood T lymphocytes and demonstrate an inhibitory effect of these novel PPAR␥ ligands produced by the monocyte 12/15-lipoxygenase on T cell IL-2 production and activity of the IL-2 promoter reporter. Furthermore, EMSA and luciferase reporter analysis revealed that the above 12/15-LO products suppressed IL-2 promoter by antagonizing the DNA binding activities and transactivation of the transcription factors NFAT and NF-B in a PPAR␥-dependent manner. Importantly, the potency of such inhibitory effects correlates well with the capability of the above metabolic lipids to activate PPAR␥. These findings suggest that activation of PPAR␥ in T cells by 12/15-LO macrophage products is a key means by which IL-4 indirectly inhibits Thp cell function.
In summary, we have identified and molecularly characterized a previously undescribed immunoregulatory circuit. Cytokines, such as IL-4, may up-regulate ligands that activate the PPAR␥ receptor expressed in T lymphocytes and exert profound indirect effects on T lymphocyte biology via nonsteroidal nuclear receptors.