Disruption of diacylglycerol metabolism impairs the induction of T cell anergy

Benjamin A Olenchock^1, 6, Rishu Guo^2, 6, Jeffery H Carpenter², Martha Jordan¹, Matthew K Topham³, Gary A Koretzky^1, 4 & Xiao-Ping Zhong^2, 5

¹ Signal Transduction Program, Abramson Family Cancer Research Institute, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104, USA.

² Department of Pediatrics, Duke University Medical Center, Durham, North Carolina 27710, USA.

³ Department of Internal Medicine, Huntsman Cancer Institute, University of Utah, Salt Lake City, Utah 84112, USA.

⁴ Department of Medicine, Division of Rheumatology, University of Pennsylvania School of Medicine, Philadelphia, Pennsylvania 19104, USA.

⁵ Department of Immunology, Duke University Medical Center, Durham, North Carolina 27710, USA.

⁶ These authors contributed equally to this work.

Anergic T cells have altered diacylglycerol metabolism, but whether that altered metabolism has a causative function in the induction of T cell anergy is not apparent. To test the importance of diacylglycerol metabolism in T cell anergy, we manipulated diacylglycerol kinases (DGKs), which are enzymes that terminate diacylglycerol-dependent signaling. Overexpression of DGK- resulted in a defect in T cell receptor signaling that is characteristic of anergy. We generated DGK--deficient mice and found that DGK--deficient T cells had more diacylglycerol-dependent T cell receptor signaling. In vivo anergy induction was impaired in DGK--deficient mice. When stimulated in anergy-producing conditions, T cells lacking DGK- or DGK- proliferated and produced interleukin 2. Pharmacological inhibition of DGK- activity in DGK--deficient T cells that received an anergizing stimulus proliferated similarly to wild-type T cells that received CD28 costimulation and prevented anergy induction. Our findings suggest that regulation of diacylglycerol metabolism is critical in determining whether activation or anergy ensues after T cell receptor stimulation.

Anergic T cells do not produce interleukin 2 (IL-2) or proliferate after reencounter with antigen, regardless of whether the secondary stimulation includes a costimulatory signal³. Anergic T cells have multiple defects in TCR signaling pathways. Compared with naive T cells, anergic T cells have larger amounts of E3 ubiquitin ligases^{5, 6}, which degrade TCR signaling proteins. In addition, anergic T cells fail to appropriately activate phospholipase C-1 or increase integrin avidity after TCR stimulation^{7, 8}. Perhaps most notably, TCR stimulation fails to activate the small GTP-binding protein Ras and its 'downstream' targets Erk1, Erk2 and AP-1 in anergic T cells^{9, 10, 11}. Also notably, diacylglycerol (DAG) analogs such as phorbol esters, which activate Ras through Ras guanyl-releasing protein 1 (RasGRP1), restore the defective activation of Ras and Erk in anergic T cells^{10, 11}.

The precise mechanism by which CD28 costimulation prevents anergy induction is not fully understood. Some experimental evidence indicates that CD28 functions in part by augmenting TCR signals and in part by transmitting distinct signals. CD28 can augment TCR-induced calcium flux, protein tyrosine kinase activity and the generation of lipid second messengers through activation of phospholipase C-1 and phosphatidylinositol-3-OH kinase^{12, 13}. CD28 induces the arginine methylation of TCR signaling proteins, activates distinct Il2 promoter elements and increases the stability of Il2 transcripts^{14, 15, 16}. CD28 promotes T cell growth and proliferation by upregulating the expression of antiapoptotic genes, by activating the phosphatidylinositol-3-OH kinase–serine-threonine kinase Akt signaling pathway and by increasing nutrient uptake and metabolism^{17, 18}. As TCR and CD28 trigger diverse yet overlapping signaling pathways, the unique signals provided by CD28 that are sufficient for anergy avoidance remain difficult to define.

Experimental evidence supports the hypothesis that DAG metabolism is key in determining whether a T cell becomes anergic or is productively activated after antigen receptor stimulation. After TCR ligation, phospholipase C-1 generates DAG and inositol trisphosphate in equimolar amounts by hydrolyzing the membrane phospholipid phosphatidylinositol-4,5-bisphosphate (PtdIns(4,5)P₂). Inositol trisphosphate acts on intracellular receptors to release calcium, and DAG links TCR stimulation to the activation of mitogen-activated protein kinase and protein kinase C. Many aspects of T cell activation are recapitulated by the pharmacological provision of calcium and DAG signals. Treatment of T cells with the calcium ionophore ionomycin and DAG analogs such as phorbol myristate acetate (PMA) is a pharmacological method commonly used to induce IL-2 production, CD69 expression, growth and proliferation while bypassing the requirement for TCR stimulation.

Treatment of T cell clones with ionomycin alone is a well established method of inducing T cell anergy^{19, 20}. The addition of PMA abrogates the ionomycin-induced anergy 'program', leading to productive T cell responses. DAG-induced activation of protein kinase C and Ras is essential for coupling TCR stimulation to the transcriptional activity of AP-1 and transcription factor NF-B. Calcium signals, in contrast, link TCR signals to the transcriptional activity of the transcription factor NF-AT. Costimulation greatly amplifies TCR-induced AP-1 and NF-B transcriptional activity^{21, 22} but has a smaller effect on NF-AT activation^{23, 24}. NF-AT activation without AP-1 activation results in the transcription of a set of anergy-associated genes, whereas cooperative NF-AT and AP-1 activity induces the transcription of genes associated with proliferation and effector functions²⁰.

Research in several laboratories has focused on how DAG signals are terminated in T cells. Phosphorylation of the DAG free hydroxyl group converts DAG into phosphatidic acid. That reaction, which is catalyzed by diacylglycerol kinases (DGKs), is important for the termination of DAG signals and for the recycling of inositol phospholipids²⁶. DGK- is an important negative regulator of TCR signaling^{27, 28}. Overexpression of DGK- in Jurkat T cells blocks TCR-induced IL-2 production (unpublished observations). TCR-induced calcium flux is unaffected by DGK- overexpression, whereas DAG-dependent signaling pathways, including Ras and AP-1 activity, are inhibited²⁸. T cells from DGK--deficient mice are hyperproliferative after TCR ligation, and that phenotype is most prominent during stimulation with small amounts of antigen.

Microarray analysis has identified the transcript encoding the -isoform of DGK (DGK-, encoded by Dgka) as being upregulated in anergic T cells and downregulated in productively activated T cells²⁰. Here we report the generation of DGK--deficient mice. We have also assessed here the importance of DAG metabolism in the induction of T cell anergy by manipulating DGK function through gene deletion, overexpression and pharmacological inhibition. Our data provide compelling evidence that regulation of DAG metabolism after TCR stimulation is critical in determining whether T cell activation or anergy will ensue.

Figure 1. Costimulation increases the hydrolysis of PtdIns(4,5)P2 and the generation of inositol trisphosphate but does not affect the conversion of DAG to phosphatidic acid. (a) Identification of 32P-labeled lipids by thin-layer chromatography. Purified, 32P-labeled CD4+ T cells were stimulated (key), then the stimulation was terminated by HCl lysis (time, horizontal axis); 32P-labeled lipids were identified by migration together with PtdIns(4,5)P2 (PIP2) standards. (b) Quantification of inositol trisphosphate (IP3). Unlabeled purified CD4+ T cells were stimulated (key), then stimulation was terminated by perchloric acid lysis (time, horizontal axis). *, P = 0.028 (paired Student's t-test). (c) Identification of 32P-labeled lipids by thin-layer chromatography as described in a; 32P-labeled lipids were identified by migration together with phosphatidic acid (PA) standards. , antibody to. Data are from four independent experiments (error bars, s.e.m.). Full Figure and legend (13K)

Evidence supporting the importance of DAG metabolism in T cell activation has come from studies demonstrating increased transcription of DGK genes in anergic cells²⁰ (T. Gajewski, personal communication), and we found that both Dgka and Dgkz were transcriptionally downregulated in activated T cells (Supplementary Fig. 1 online). To test more directly how DAG metabolism affects T cell responses, we manipulated DGK activity both genetically and pharmacologically. Overexpression of DGK- in Jurkat T cells impairs DAG-dependent TCR signaling but does not affect inositol trisphosphate–induced calcium flux²⁸. That TCR signaling abnormality is reminiscent of that seen in anergic primary T cells, which fail to link TCR signals to AP-1 transcriptional activity⁹. As DGK- is also highly expressed in and is upregulated in anergic T cells²⁰ (T. Gajewski, personal communication), we sought to determine whether DGK- overexpression resulted in a similar signaling alteration. Overexpressed DGK- greatly impaired TCR-induced activation of a cotransfected AP-1-driven luciferase reporter construct (Fig. 2a). TCR-induced calcium flux, however, was unaffected by DGK- overexpression (Fig. 2b).

Figure 2. Forced DGK- expression in Jurkat T cells blocks TCR-induced AP-1 activity but does not affect calcium flux. Jurkat T cells were transiently transfected with a vector encoding DGK- or empty vector (pCMV) plus either an AP-1-responsive luciferase construct (a) or a GFP construct (b) to label transfected cells. (a) Luciferase activity in lysates of transfected cells left unstimulated or stimulated for 8 h with the TCR-specific monoclonal antibody c305 (dilution, 1:30,000 (1:30K) or 1:60,000 (1:60K)) or with PMA plus ionomycin (PMA + Io). (b) Flow cytometry of calcium flux in GFP+ Jurkat cells transfected with expression vectors (key), labeled with the calcium-sensitive dye Indo-1 and stimulated with various dilutions of c305 (data represent a dilution of 1:480,000). Data are representative of two experiments (error bars (a), s.e.m.). Full Figure and legend (21K)

To complement the gain-of-function studies in Jurkat cells and to extend our studies to primary cells, we generated Dgka^-/- mice (Supplementary Fig. 2 online). We replaced a 3.3-kilobase genomic region of the Dgka locus, including exons encoding the C1B domain and the ATP-binding site of the kinase domain, with a neomycin-resistance cassette. We demonstrated successful gene deletion by Southern blot and PCR analysis (Supplementary Fig. 2). Dgka^-/- mice were born at the expected mendelian ratio and seemed normal. Assessment of thymocyte numbers and populations showed no gross alteration in thymocyte development (Supplementary Fig. 3 online). The numbers and percentages of CD4⁺CD8⁺, CD4⁺, CD8⁺ and CD4^-CD8^- thymocyte populations in Dgka^-/- mice were similar to those in wild-type littermates. Peripheral T cells in Dgka^-/- mice were present in normal numbers and percentages in the spleen and lymph node (Fig. 3a). Naive populations (CD62L^hiCD44^lo) and effector and memory populations (CD62L^loCD44^hi) were present in the appropriate ratios (Fig. 3b and Supplementary Fig. 4 online), and basal TCR and CD69 expression on CD4⁺ T cells (Fig. 3c) and CD8⁺ T cells (Supplementary Fig. 4) was unaffected by DGK- deficiency. We bred Dgka^-/- mice with Dgkz^-/- mice to create mice lacking both DGK isoforms. Dgka^-/-Dgkz^-/- mice were viable but had substantial defects in thymocyte development (R.G. and X.-P.Z., unpublished observations). However, we detected a few CD4⁺ and CD8⁺ thymocytes and peripheral T cells. Those cells had constitutive expression of activation markers, and preliminary data indicated that immune cells in Dgka^-/-Dgkz^-/- mice were inappropriately activated (R.G. and X.-P.Z., unpublished observations). Thus, study of resting Dgka^-/-Dgkz^-/- T cells is not feasible at present. After TCR stimulation, Dgka^-/- T cells generated phosphatidic acid in amounts near those produced by wild-type T cells, although there was a consistent trend toward diminished phosphatidic acid production in Dgka^-/- T cells (data not shown). That finding contrasts with the more notable effect of DGK- deficiency on TCR-induced phosphatidic acid production²⁷, suggesting differences in activity or substrate pool for these enzymes.

Figure 3. DGK- deficiency does not alter the surface phenotype of peripheral CD4+ T cells. Flow cytometry of splenic and lymph node cells from wild-type and Dgka-/- mice, stained with fluorescence-labeled antibodies (along margins). (a) All live cells. (b,c) Gated on CD4+ T cells. Numbers in quadrants indicate percent cells in each. Data are representative of five experiments. Full Figure and legend (36K)

We next assessed whether DAG-dependent TCR signaling was altered by DGK- deficiency. We stimulated purified Dgka^-/- T cells with antibody to CD3 (anti-CD3) and measured Ras activity and Erk phosphorylation. TCR-induced Ras activity was enhanced in Dgka^-/- T cells (Fig. 4a) and Erk phosphorylation was prolonged (Fig. 4b). The effects of DGK- deficiency on Ras and Erk activity were reproducible but, notably, were less than those in Dgkz^-/- T cells²⁷. As with DGK- deficiency, TCR-induced calcium flux was unaffected by DGK- deficiency (Fig. 4c). Thus, DGK deficiency seems to selectively augment DAG-dependent TCR signals.

Figure 4. DGK- deficiency results in increased DAG-dependent TCR signaling but does not affect calcium flux. (a,b) Purified splenic T cells from wild-type mice (WT) and Dgka-/- mice were left unstimulated (0) or were stimulated with TCR-specific antibody (times, above lanes) or with PMA (P). (a) Immunoblot quantification of GTP-bound Ras precipitated with glutathione S-transferase–Raf—Ras-binding domain beads and analyzed with an antibody specific for Ras. KO, Dgka-/-. (b) Immunoblot of Erk activity in stimulated T cells (top), analyzed with an antibody specific for phosphorylated Erk (p-Erk) or total Erk (Erk 1/2; loading control). Bottom, quantification of immunoblot band intensities; phosphorylated Erk signals are normalized to those of total Erk. (c) Flow cytometry of calcium flux in wild-type and Dgka-/- CD4+ thymocytes labeled at 30 °C with fluorescent anti-CD4 and anti-CD8, the calcium-sensitive dye Indo-1 and unlabeled anti-CD3 (2C11); TCR crosslinking was induced by the addition of the secondary antibody goat anti-hamster (gh) at 37 °C. Data are representative of three (a,b) or five (c) experiments. Full Figure and legend (24K)

Figure 5. Dgka-/- T cells are hyperproliferative. Lymph node and splenic cells from wild-type and Dgka-/- mice were stimulated (horizontal axes) for 64 h and then were pulsed with [3H]thymidine; proliferation was assessed 8 h later by scintillation counting. Data are representative of three experiments. Full Figure and legend (15K)

We hypothesized that enhanced TCR-induced DAG signaling after the addition of CD28 costimulation contributed to productive T cell activation and anergy avoidance. To test that hypothesis with a genetic approach, we first sought to determine whether DGK- deficiency, which augmented DAG-dependent signaling, would impair T cell anergy induction in vivo. We injected wild-type and Dgka^-/- mice with the 'superantigen' staphylococcal enterotoxin B (SEB), which renders V8⁺ T cells refractory to restimulation with SEB (inducing T cell anergy)^{31, 32}. After injection of SEB, wild-type and Dgka^-/- splenocytes contained a similar percentage of V8⁺ T cells, indicating that DGK- deficiency did not alter clonotype selection or superantigen-mediated deletion (Fig. 6a). When restimulated with SEB ex vivo, wild-type T cells produced very little IL-2 and showed blunted proliferation (Fig. 6b). In contrast, Dgka^-/- T cells from SEB-injected mice retained the ability to produce IL-2 and to proliferate in response to SEB restimulation (Fig. 6b). These data indicated that DGK- is essential for anergy induction in vivo.

Figure 6. In vivo anergy induction is impaired by DGK- deficiency. Wild-type and Dgka-/- mice were injected intraperitoneally with PBS (data not shown) or 100 g SEB; splenocytes were collected 7 d later. (a) Flow cytometry to determine the percent V8+ and V10+ splenocytes in SEB-treated mice. (b) ELISA of IL-2 in supernatants and quantification of proliferation by [3H]thymidine incorporation for splenocytes from SEB-treated mice restimulated with SEB or PMA plus ionomycin (P+I). None, no restimulation. Data are representative of three experiments (error bars, s.e.m.). Full Figure and legend (20K)

Next we assessed the responses of Dgka^-/- primary T cells to TCR stimulation without costimulation. As noted above, intrathymic T cell development in Dgka^-/-Dgkz^-/- mice was grossly abnormal, precluding the evaluation of anergy in Dgka^-/-Dgkz^-/- T cells. Therefore, as an alternative approach to examine the effects of combined deficiency of DGK- and DGK- on anergy induction, we used the type I DGK inhibitor R59022 to inhibit DGK- function in wild-type or Dgkz^-/- T cells³³. We depleted wild-type, Dgkz^-/- and Dgka^-/- splenocyte samples of CD8⁺ cells and stimulated them with anti-CD3 in the presence of a fusion protein of cytotoxic T lymphocyte antigen 4 and Fc (CTLA-4–Fc) to block CD28 costimulation. We monitored cell proliferation by dilution of carboxyfluoresein succinimidyl ester (CFSE). At 48 h after stimulation, very few surviving wild-type T cells divided (Fig. 7a and Supplementary Fig. 5 online). In contrast, Dgkz^-/- and Dgka^-/- T cells underwent two to three rounds of cell division. Pharmacological inhibition of DGK- in wild-type T cells resulted in a 'phenocopy' of DGK- deficiency (Fig. 7a), and in Dgkz^-/- T cells, it augmented the proliferative response (Fig. 7a). Dgkz^-/- T cells treated with the pharmacological inhibitor of DGK- grew and divided like wild-type T cells that received CD28 costimulation (Fig. 7b).

Figure 7. DGK deficiency decreases the requirement for costimulation. Splenocyte samples from wild-type and Dgkz-/- mice were depleted of CD8+ cells, were labeled with CFSE and were incubated in the presence or absence of the DGK- inhibitor R59022. Cells were stimulated with anti-CD3 (2C11) and CTLA-4–Fc (to block costimulation) or with anti-CD28 (to provide costimulation). (a) Flow cytometry of CFSE dilution in gated CD4+ T cells stimulated with anti-CD3 and CTLA-4–Fc. (b) Flow cytometry of CFSE dilution in gated CD4+ T cells stimulated with anti-CD3, anti-CD28 and/or CTLA-4–Fc (key). (c) ELISA of IL-2 in supernatants of cells stimulated with anti-CD3, anti-CD28 and/or CTLA-4–Fc (below graph). Data are representative of four experiments (error bars (c), s.e.m.). Full Figure and legend (28K)

As DAG and calcium signals combine to drive transcription of IL-2, we next determined whether the enhanced proliferation of DGK-deficient T cells was associated with increased IL-2 production. As expected, wild-type cells stimulated in conditions resulting in anergy produced very little or no IL-2 (Fig. 7c). Dgka^-/- and Dgkz^-/- T cells produced more IL-2 after stimulation with anti-CD3 and CTLA-4–Fc, but wild-type T cells treated with R59022 produced even more IL-2. The difference between pharmacological and genetic blockade of DGK- function probably reflects some nonspecific DGK inhibition by R59022 (ref. 33), but nonetheless all 'singly deficient' cells produced little IL-2 compared with wild-type cells that received costimulation (Fig. 7c). However, production of IL-2 by Dgkz^-/- T cells treated with the DGK- inhibitor was near that of wild-type cells that received CD28 costimulation (Fig. 7c).

To further investigate the function of DGKs in T cell anergy, we measured cell growth and IL-2 production of 'anergized' wild-type and DGK-deficient T cells in response to restimulation with anti-CD3 and anti-CD28 (Fig. 8). We assessed cell growth by measuring forward scatter of light by flow cytometry and quantified IL-2 production by enzyme-linked immunosorbent assay (ELISA). Because of the substantial cell death that occurs during anergy induction, cell yield among anergy experiments was variable. Therefore, to normalize data among experiments, we assessed IL-2 production by T cells from experimental mice relative to that of naive wild-type T cells. Wild-type T cells stimulated with anti-CD3 and CTLA-4–Fc made very little IL-2 and showed impaired growth in response to restimulation with anti-CD3 and anti-CD28 (Fig. 8); thus, they acted like anergic T cells. In contrast, stimulation of Dgka^-/- or Dgkz^-/- T cells with anti-CD3 and CTLA-4–Fc did not result in anergy, as these T cells retained the ability to grow and produce IL-2 after restimulation. Notably, pharmacological inhibition of DGK- during the primary anergy-producing stimulation had a less pronounced effect on anergy induction than did DGK- deficiency, as cells that underwent the former treatment showed blunted growth responses and produced less IL-2 after restimulation. Dgkz^-/- T cells treated with the inhibitor of DGK- produced nearly as much IL-2 after restimulation as did naive wild-type T cells but not as much as previously activated wild-type T cells did (Fig. 8 and data not shown). These data collectively demonstrated that enhancing DAG-dependent signaling by DGK deficiency impairs anergy induction.

Figure 8. DGK deficiency impairs T cell anergy. After 72 h of primary stimulation with anti-CD3 and CTLA-4–Fc, CD4+ T cells were isolated by positive selection, were allowed to 'rest' overnight and then were replated onto a monolayer of bone marrow–derived macrophages; cells were left unstimulated or were stimulated overnight with anti-CD3 and anti-CD28. Naive, CD4+ T cells purified from the spleen of a wild-type mouse. Left, flow cytometry of forward scatter (FSC). Right, ELISA of IL-2 production, normalized to IL-2 production from naive CD4+ T cells (100%, vertical dashed line). Data are representative of three independent experiments (error bars (right), s.e.m.). Full Figure and legend (25K)

	Top

Here we have used overexpression as well as genetic and pharmacological inhibition of DGK to examine the importance of DAG metabolism in anergy induction. In agreement with published data²⁹, we have demonstrated that the hydrolysis of PtdIns(4,5)P₂ into inositol trisphosphate was augmented by CD28 costimulation. The conversion of DAG to phosphatidic acid, however, was unaffected, indicating that costimulation increases DAG accumulation. Inhibition of DAG metabolism by DGK overexpression resulted in impaired TCR-induced AP-1 activity, a signaling defect similar to that of anergic T cells. As a complementary approach, we generated Dgka^-/- and Dgkz^-/- mice. We predicted that by impairing DAG removal through DGK deficiency, stimulation with TCR alone would fail to make T cells anergic. In agreement with that hypothesis, in vivo anergy induction was greatly impaired in Dgka^-/- mice. Dgka^-/- and Dgkz^-/- T cells showed enhanced proliferation and IL-2 production after anergy-producing stimulation and retained the ability to produce IL-2 after restimulation. In addition, in response to anergy-producing stimuli, the proliferation and IL-2 production of Dgkz^-/- T cells treated with a pharmacological inhibitor of DGK- was similar to that of wild-type T cells that received CD28 costimulation.

Enhanced TCR-induced proliferation of DGK-deficient T cells correlated with enhanced IL-2 production. However, IL-2 production in 'singly deficient' cells was not substantial, and in some experiments, Dgkz^-/- T cells treated with the DGK- inhibitor produced less IL-2 than did costimulated wild-type cells but proliferated to a similar extent. There is evidence from the IL-2-dependent cell line CTLL-2 that DGK enzymes regulate IL-2 responsiveness^{35, 36, 37, 38, 39}. In those studies, it was proposed that phosphatidic acid produced by DGK- is necessary for IL-2-dependent cell cycle progression, as inhibition of DGK- blunted proliferation in response to IL-2. We do not yet know whether the increased proliferation of DGK-deficient T cells in response to anergy-producing stimuli is due entirely to increased IL-2 production or whether we unexpectedly increased IL-2 responsiveness by removing DGK function.

By impairing DAG metabolism, we abrogated the induction of T cell anergy. One possible mechanism by which that was accomplished was through increased IL-2 production and proliferation. There is evidence that proliferation is closely linked to anergy 'avoidance', perhaps by the dilution of cell cycle inhibitors, including p27^kip1 (refs. 40,41). Although Dgkz^-/- T cells treated with the inhibitor of DGK- did produce IL-2 after restimulation, they produced much less than did previously activated effector T cells (data not shown). CD28 costimulation affects multiple TCR signaling pathways in addition to increasing DAG accumulation. We surmise that by increasing DAG-dependent signaling, we increased AP-1 transcriptional activity, which could 'cooperate' with NF-AT and block the induction of anergy-associated genes²⁰. Further work is needed to explore the biochemical alterations involved in those increases in IL-2 production and proliferation.

The possibility that diminished phosphatidic acid production contributed to the increased responsiveness and anergy 'avoidance' of DGK-deficient T cells cannot be excluded. Phosphatidic acid is a bioactive lipid second messenger with many purported functions, including regulation of the mammalian target of rapamycin, cytoskeletal alterations and membrane trafficking^{42, 43, 44}. Phosphatidic acid is also involved in phosphatidylinositol recycling, and it can affect the generation of PtdIns(4,5)P₂ and the activity of phospholipase C-1 and phosphatidylinositol-3-OH kinase^{45, 46, 47}. Additionally, there are many pathways to phosphatidic acid production, including the hydrolysis of phosphatidylcholine and acylation of lyso-PA by phospholipase D⁴³. Further studies are needed explore the effect of impaired TCR-induced phosphatidic acid production on activation of the mammalian target of rapamycin and cell survival.

T cells have detectable expression of transcripts encoding at least three DGK isoforms: DGK-, DGK- and DGK- (unpublished observations). Our overexpression studies and analyses of DGK-deficient T cells have not demonstrated a functional difference between DGK- and DGK-, although the signaling alterations in Dgkz^-/- T cells seemed to be greater than those in Dgkz^-/- T cells^{27, 28}. We suspect that the activities of these DGK isoforms might be differentially regulated in T cells, as they have structurally distinct regulatory domains and differ in their subcellular localization²⁶. DGK- is recruited to the plasma membrane after receptor stimulation^{34, 48}, but recruitment of DGK- to the plasma membrane after TCR stimulation of Jurkat or primary T cells has not been noted⁴⁹ (unpublished observations). In other cell types, DGK- localizes to the nucleus or the plasma membrane, where it regulates DAG metabolism^{50, 51, 52, 53}. Furthermore, DGK- activity can be regulated by calcium, tyrosine phosphorylation or inositol trisphosphate, whereas DGK- localization and function is regulated by protein kinase C^{49, 52, 54, 55}. Membranes differ in lipid composition, and although DGK- and DGK- show little substrate specificity for diacylglycerol isoforms in vitro, substrate accessibility could determine in vivo specificity. Additionally, it is possible that DGK- and DGK- have kinase-independent functions in T cells, and at present we cannot exclude the possibility of a contribution of those DAG-independent effects.

DGK- and DGK- are transcriptionally regulated in T cells, with more transcript and protein in anergic T cells and less transcript in activated T cells²⁰ (T. Gajewski, personal communication). In addition, a chromatin modification associated with active gene transcription is present at high density at the Dgkz locus in naive cells and is rapidly lost after TCR stimulation⁵⁶. Those data demonstrate that DGK expression is regulated in T cells and correlates with the threshold for T cell activation.

	Top

CD4⁺ T cells were isolated from spleens by positive selection with MACS columns (Militenyi) and were allowed to 'rest' for 30 min at 37 °C in Tyrode's buffer (10 mM HEPES, pH 7.4, 130 mM NaCl, 1 mM MgCl₂, 5 mM KCl, 1.4 mM CaCl₂, 5.6 mM glucose and 1 mg/ml of BSA). Splenocyte samples depleted of T cells (isolated by negative selection by removing CD4⁺ and CD8⁺ cells; Militenyi) were used as antigen-presenting cells (APCs). APCs (20 10⁶ to 30 10⁶) were incubated for 30 min at 37 °C with anti-CD3 (1 g/ml; 145-2C11; BD Pharmingen) along with anti-CD28 (0.5 g/ml; 37.51; BD Pharmingen) or CTLA-4–Fc (5 g/ml; R&D Systems) in Tyrode's buffer. Cells were washed once in Tyrode's buffer, and then purified CD4⁺ T cells (8 10⁶ to 10 10⁶) were added to the APCs, followed by centrifugation at 3000 r.p.m. in a Sorvall Biofuge 'pico' centrifuge for 15 s to initiate T cell–APC contact. Stimulation was terminated by incubation for 20 min on ice with 0.2 volumes of 20% perchloric acid. Proteins were sedimented, and supernatants were neutralized to a pH of 7.5 by titration of 1.5 M KOH and 60 mM HEPES containing universal indicator dye (Sigma). KClO₄ was precipitated by centrifugation, and inositol trisphosphate was measured in supernatants by radioreceptor assay according to the manufacturer's protocol (PerkinElmer).

Purified CD4⁺ T cells were 'starved' of phosphate for 1 h in Tyrode's buffer before stimulation, followed by supplementation for 1 h with 0.25 mCi/ml [³²P]orthophosphate (MP Biomedicals). Cells were stimulated with anti-CD3 (5 g/ml; 500A2; BD Pharmingen) with or without anti-CD28 (0.5 mg/ml) and were lysed by the addition of 0.1 volumes of 2 M HCl. Lipids were extracted and were analyzed by thin-layer chromatography in a basic solvent (9:7:2 chloroform:methanol:4 M NH₄OH). Phosphatidic acid and PtdIns(4,5)P₂ were identified by migration compared with that of commercial standards (Avanti Polar Lipids).

Jurkat T cells were transiently transfected with 20 g of the Dgka construct and 10 g of a GFP construct (MSCV-IRES-GFP; MigR1) for labeling of transfected cells or with GFP alone as a control. Calcium flux was measured as described²⁸. Additional methods descriptions are in the Supplementary Methods.

Purified splenic T cells were stimulated for various times with 5 g/ml of anti-CD3 (500A2; BD Pharmingen) and were lysed in 1% Nonidet P-40 lysis buffer (1% (volume/volume) Nonidet-40, 150 mM NaCl and 50 mM Tris, pH 7.4) with protease inhibitors. Proteins were resolved by SDS-PAGE and were transferred to a Trans-Blot Nitrocellulose membrane (Bio-Rad Laboratories); membranes were probed with antibodies specific to phosphorylated Erk (91015; Cell Signal Technology) and phospholipase C-1 (05-163; Upstate Biotechnology). Membranes were stripped and were reprobed for analysis of total Erk (SC-16982; Santa Cruz Biotechnology). Activated Ras in cell lysates was determined by glutathione S-transferase–Raf—Ras-binding domain precipitation assay as described²⁸.

Wild-type or Dgka^-/- mice were injected intraperitoneally with PBS or with 100 g SEB (Sigma) in PBS. Then, 7 d later, mice were killed and single-cell suspensions were made from spleens. Some cells were stained with anti-CD8, anti-CD4, anti-V8 (MR5-2) and anti-V10 (B21.5; both BD Pharmigen); the remaining splenocytes were plated in U-bottomed 96-well plates at a density of 5 10⁵ cells per well in 200 l RPMI-10 medium. Cells were left unstimulated or were stimulated with SEB (10 g/ml) or with PMA (50 ng/ml) plus ionomycin (200 ng/ml). At 40 h after stimulation, IL-2 in culture supernatants was measured by ELISA. Cells were pulsed with [³H]thymidine for another 8 h for measurement of T cell proliferation.

Wild-type, Dgka^-/- and Dgkz^-/- splenocytes were stained with 5 M CFSE, were stimulated for 72 h with anti-CD3 (1 g/ml; 2C11) along with CTLA-4–Fc (5 g/ml), were stained with allophycocyanin-conjugated anti-CD4 and were analyzed by flow cytometry. Cell division was assessed by CFSE dilution after gating on live CD4⁺ cells. Alternatively, cells were stimulated for 72 h and were pulsed with 1 Ci/well of [³H]thymidine for the final 8 h of stimulation, and proliferation was assessed by tritium incorporation with a scintillation counter. For restimulation analyses, cells were prestimulated with anti-CD3 plus CTLA-4–Fc, then after 72 h, CD4⁺ cells were purified by negative selection (with fluorescein isothiocyanate–conjugated anti-CD8, anti-B220 (RA3-6B2; BD Pharmingen), anti-DX5 and anti-CD11b (M1/70; BD Pharmingen), followed by depletion with anti–fluorescein isothiocyanate magnetic beads) and were allowed to 'rest' overnight at 37 °C. Live cells were then counted by Trypan blue exclusion, and equivalent numbers of live cells were dropped onto monolayers of bone marrow–derived macrophages coated with anti-CD3 (1 g/ml) and anti-CD28 (0.5 g/ml). After 24 h, supernatants were collected and IL-2 was quantified by ELISA according to the manufacturer's protocol (R&D Systems).

	Top

1. Mathis, D. & Benoist, C. Back to central tolerance. Immunity 20, 509–516 (2004). | Article | PubMed | ISI | ChemPort |
2. Bouneaud, C. , Kourilsky, P. & Bousso, P. Impact of negative selection on the T cell repertoire reactive to a self-peptide: a large fraction of T cell clones escapes clonal deletion. Immunity 13, 829–840 (2000). | Article | PubMed | ISI | ChemPort |
3. Schwartz, R.H. T cell anergy. Annu. Rev. Immunol. 21, 305–334 (2003). | Article | PubMed | ISI | ChemPort |
4. Kopp, E. & Medzhitov, R. Recognition of microbial infection by Toll-like receptors. Curr. Opin. Immunol. 15, 396–401 (2003). | Article | PubMed | ISI | ChemPort |
5. Heissmeyer, V. et al. Calcineurin imposes T cell unresponsiveness through targeted proteolysis of signaling proteins. Nat. Immunol. 5, 255–265 (2004). | Article | PubMed | ISI | ChemPort |
6. Safford, M. et al. Egr-2 and Egr-3 are negative regulators of T cell activation. Nat. Immunol. 6, 472–480 (2005). | Article | PubMed | ChemPort |
7. Gajewski, T.F. , Qian, D. , Fields, P. & Fitch, F.W. Anergic T-lymphocyte clones have altered inositol phosphate, calcium, and tyrosine kinase signaling pathways. Proc. Natl. Acad. Sci. USA 91, 38–42 (1994). | Article | PubMed | ChemPort |
8. Wells, A.D. et al. Regulation of T cell activation and tolerance by phospholipase C -1-dependent integrin avidity modulation. J. Immunol. 170, 4127–4133 (2003). | PubMed | ISI | ChemPort |
9. Kang, S.M. et al. Transactivation by AP-1 is a molecular target of T cell clonal anergy. Science 257, 1134–1138 (1992). | PubMed | ISI | ChemPort |
10. Li, W. , Whaley, C.D. , Mondino, A. & Mueller, D.L. Blocked signal transduction to the ERK and JNK protein kinases in anergic CD4⁺ T cells. Science 271, 1272–1276 (1996). | PubMed | ISI | ChemPort |
11. Fields, P.E. , Gajewski, T.F. & Fitch, F.W. Blocked Ras activation in anergic CD4⁺ T cells. Science 271, 1276–1278 (1996). | PubMed | ISI | ChemPort |
12. Acuto, O. , Mise-Omata, S. , Mangino, G. & Michel, F. Molecular modifiers of T cell antigen receptor triggering threshold: the mechanism of CD28 costimulatory receptor. Immunol. Rev. 192, 21–31 (2003). | Article | PubMed | ChemPort |
13. Frauwirth, K.A. & Thompson, C.B. Regulation of T lymphocyte metabolism. J. Immunol. 172, 4661–4665 (2004). | PubMed | ISI | ChemPort |
14. Lindstein, T. , June, C.H. , Ledbetter, J.A. , Stella, G. & Thompson, C.B. Regulation of lymphokine messenger RNA stability by a surface-mediated T cell activation pathway. Science 244, 339–343 (1989). | PubMed | ChemPort |
15. Blanchet, F. , Cardona, A. , Letimier, F.A. , Hershfield, M.S. & Acuto, O. CD28 costimulatory signal induces protein arginine methylation in T cells. J. Exp. Med. 202, 371–377 (2005). | Article | PubMed | ISI | ChemPort |
16. Shapiro, V.S. , Truitt, K.E. , Imboden, J.B. & Weiss, A. CD28 mediates transcriptional upregulation of the interleukin-2 (IL-2) promoter through a composite element containing the CD28RE and NF-IL-2B AP-1 sites. Mol. Cell. Biol. 17, 4051–4058 (1997). | PubMed | ISI | ChemPort |
17. Cory, S. & Adams, J.M. The Bcl2 family: regulators of the cellular life-or-death switch. Nat. Rev. Cancer 2, 647–656 (2002). | Article | PubMed | ISI | ChemPort |
18. Fox, C.J. , Hammerman, P.S. & Thompson, C.B. Fuel feeds function: energy metabolism and the T-cell response. Nat. Rev. Immunol. 5, 844–852 (2005). | Article | PubMed | ChemPort |
19. Jenkins, M.K. , Pardoll, D.M. , Mizuguchi, J. , Chused, T.M. & Schwartz, R.H. Molecular events in the induction of a nonresponsive state in interleukin 2-producing helper T-lymphocyte clones. Proc. Natl. Acad. Sci. USA 84, 5409–5413 (1987). | Article | PubMed | ChemPort |
20. Macian, F. et al. Transcriptional mechanisms underlying lymphocyte tolerance. Cell 109, 719–731 (2002). | Article | PubMed | ISI | ChemPort |
21. Su, B. et al. JNK is involved in signal integration during costimulation of T lymphocytes. Cell 77, 727–736 (1994). | Article | PubMed | ISI |
22. Harhaj, E.W. & Sun, S.C. IB kinases serve as a target of CD28 signaling. J. Biol. Chem. 273, 25185–25190 (1998). | Article | PubMed | ISI | ChemPort |
23. Lyakh, L. , Ghosh, P. & Rice, N.R. Expression of NFAT-family proteins in normal human T cells. Mol. Cell. Biol. 17, 2475–2484 (1997). | PubMed | ISI | ChemPort |
24. Dolmetsch, R.E. , Lewis, R.S. , Goodnow, C.C. & Healy, J.I. Differential activation of transcription factors induced by Ca²⁺ response amplitude and duration. Nature 386, 855–858 (1997). | Article | PubMed | ISI | ChemPort |
25. Dower, N.A. et al. RasGRP is essential for mouse thymocyte differentiation and TCR signaling. Nat. Immunol. 1, 317–321 (2000). | Article | PubMed | ISI | ChemPort |
26. Luo, B. , Regier, D.S. , Prescott, S.M. & Topham, M.K. Diacylglycerol kinases. Cell. Signal. 16, 983–989 (2004). | Article | PubMed | ISI | ChemPort |
27. Zhong, X.P. et al. Enhanced T cell responses due to diacylglycerol kinase- deficiency. Nat. Immunol. 4, 882–890 (2003). | Article | PubMed | ISI | ChemPort |
28. Zhong, X.P. et al. Regulation of T cell receptor-induced activation of the Ras-ERK pathway by diacylglycerol kinase . J. Biol. Chem. 277, 31089–31098 (2002). | Article | PubMed | ISI | ChemPort |
29. Berney, S.M. & Atkinson, T.P. Phosphatidylinositol hydrolysis in freshly isolated human T lymphocytes. J. Immunol. Methods 186, 71–77 (1995). | Article | PubMed | ChemPort |
30. Antony, P. et al. B-cell antigen receptor activates transcription factors NFAT (nuclear factor of activated T-cells) and NF-B (nuclear factor B) via a mechanism that involves diacylglycerol. Biochem. Soc. Trans. 32, 113–115 (2004). | Article | PubMed | ChemPort |
31. Rellahan, B.L. , Jones, L.A. , Kruisbeek, A.M. , Fry, A.M. & Matis, L.A. In vivo induction of anergy in peripheral V8⁺ T cells by staphylococcal enterotoxin B. J. Exp. Med. 172, 1091–1100 (1990). | Article | PubMed | ISI | ChemPort |
32. Kawabe, Y. & Ochi, A. Selective anergy of V8⁺,CD4⁺ T cells in staphylococcus enterotoxin B-primed mice. J. Exp. Med. 172, 1065–1070 (1990). | Article | PubMed | ChemPort |
33. Jiang, Y. , Sakane, F. , Kanoh, H. & Walsh, J.P. Selectivity of the diacylglycerol kinase inhibitor 3-[2-(4-[bis-(4-fluorophenyl)methylene]-1-piperidinyl)ethyl]-2, 3-dihydro-2-thioxo-4(1H)quinazolinone (R59949) among diacylglycerol kinase subtypes. Biochem. Pharmacol. 59, 763–772 (2000). | Article | PubMed | ISI | ChemPort |
34. Jones, D.R. , Sanjuan, M.A. , Stone, J.C. & Merida, I. Expression of a catalytically inactive form of diacylglycerol kinase induces sustained signaling through RasGRP. FASEB J. 16, 595–597 (2002). | PubMed | ChemPort |
35. Merida, I. , Williamson, P. , Smith, K. & Gaulton, G.N. The role of diacylglycerol kinase activation and phosphatidate accumulation in interleukin-2-dependent lymphocyte proliferation. DNA Cell Biol. 12, 473–479 (1993). | PubMed | ChemPort |
36. Williamson, P. , Merida, I. & Gaulton, G. Protein and lipid kinase activation cascades in interleukin-2 receptor signalling. Semin. Immunol. 5, 337–344 (1993). | Article | PubMed | ChemPort |
37. Flores, I. , Casaseca, T. , Martinez, A.C. , Kanoh, H. & Merida, I. Phosphatidic acid generation through interleukin 2 (IL-2)-induced -diacylglycerol kinase activation is an essential step in IL-2-mediated lymphocyte proliferation. J. Biol. Chem. 271, 10334–10340 (1996). | Article | PubMed | ISI | ChemPort |
38. Jones, D.R. , Flores, I. , Diaz, E. , Martinez, C. & Merida, I. Interleukin-2 stimulates a late increase in phosphatidic acid production in the absence of phospholipase D activation. FEBS Lett. 433, 23–27 (1998). | Article | PubMed | ChemPort |
39. Flores, I. et al. Diacylglycerol kinase inhibition prevents IL-2-induced G1 to S transition through a phosphatidylinositol-3 kinase-independent mechanism. J. Immunol. 163, 708–714 (1999). | PubMed | ISI | ChemPort |
40. Boonen, G.J. et al. CD28 induces cell cycle progression by IL-2-independent down-regulation of p27^kip1 expression in human peripheral T lymphocytes. Eur. J. Immunol. 29, 789–798 (1999). | Article | PubMed | ISI | ChemPort |
41. Jenkins, M.K. The role of cell division in the induction of clonal anergy. Immunol. Today 13, 69–73 (1992). | Article | PubMed | ChemPort |
42. Topham, M.K. & Prescott, S.M. Mammalian diacylglycerol kinases, a family of lipid kinases with signaling functions. J. Biol. Chem. 274, 11447–11450 (1999). | Article | PubMed | ISI | ChemPort |
43. English, D. Phosphatidic acid: a lipid messenger involved in intracellular and extracellular signalling. Cell. Signal. 8, 341–347 (1996). | Article | PubMed | ISI | ChemPort |
44. Avila-Flores, A. , Santos, T. , Rincon, E. & Merida, I. Modulation of the mammalian target of rapamycin pathway by diacylglycerol kinase-produced phosphatidic acid. J. Biol. Chem. 280, 10091–10099 (2005). | Article | PubMed | ChemPort |
45. Jones, G.A. & Carpenter, G. The regulation of phospholipase C-1 by phosphatidic acid. Assessment of kinetic parameters. J. Biol. Chem. 268, 20845–20850 (1993). | PubMed | ISI | ChemPort |
46. Luo, B. , Prescott, S.M. & Topham, M.K. Diacylglycerol kinase regulates phosphatidylinositol 4-phosphate 5-kinase I by a novel mechanism. Cell. Signal. 16, 891–897 (2004). | Article | PubMed | ChemPort |
47. Lauener, R. , Shen, Y. , Duronio, V. & Salari, H. Selective inhibition of phosphatidylinositol 3-kinase by phosphatidic acid and related lipids. Biochem. Biophys. Res. Commun. 215, 8–14 (1995). | Article | PubMed | ChemPort |
48. Sanjuan, M.A. et al. T cell activation in vivo targets diacylglycerol kinase to the membrane: a novel mechanism for Ras attenuation. J. Immunol. 170, 2877–2883 (2003). | PubMed | ISI | ChemPort |
49. Fukunaga-Takenaka, R. et al. Importance of chroman ring and tyrosine phosphorylation in the subtype-specific translocation and activation of diacylglycerol kinase by D--tocopherol. Genes Cells 10, 311–319 (2005). | Article | PubMed | ChemPort |
50. Ali, H. et al. Selective translocation of diacylglycerol kinase in hippocampal neurons under transient forebrain ischemia. Neurosci. Lett. 372, 190–195 (2004). | Article | PubMed | ChemPort |
51. D'Santos, C.S. , Clarke, J.H. , Irvine, R.F. & Divecha, N. Nuclei contain two differentially regulated pools of diacylglycerol. Curr. Biol. 9, 437–440 (1999). | Article | PubMed | ChemPort |
52. Topham, M.K. et al. Protein kinase C regulates the nuclear localization of diacylglycerol kinase-. Nature 394, 697–700 (1998). | Article | PubMed | ISI | ChemPort |
53. Abramovici, H. , Hogan, A.B. , Obagi, C. , Topham, M.K. & Gee, S.H. Diacylglycerol kinase-zeta localization in skeletal muscle is regulated by phosphorylation and interaction with syntrophins. Mol. Biol. Cell 14, 4499–4511 (2003). | Article | PubMed | ChemPort |
54. Cipres, A. et al. Regulation of diacylglycerol kinase by phosphoinositide 3-kinase lipid products. J. Biol. Chem. 278, 35629–35635 (2003). | Article | PubMed | ChemPort |
55. Luo, B. , Prescott, S.M. & Topham, M.K. Protein kinase C phosphorylates and negatively regulates diacylglycerol kinase . J. Biol. Chem. 278, 39542–39547 (2003). | Article | PubMed | ISI | ChemPort |
56. Vakoc, C.R. , Mandat, S.A. , Olenchock, B.A. & Blobel, G.A. Histone H3 lysine 9 methylation and HP1 are associated with transcription elongation through mammalian chromatin. Mol. Cell 19, 381–391 (2005). | Article | PubMed | ISI | ChemPort |

	Top

Competing interests statement:  The authors declare that they have no competing financial interests.