Nuclear Diacylglycerol Kinase-θ Is Activated in Response to α-Thrombin*

Currently, there is substantial evidence that nuclear lipid metabolism plays a critical role in a number of signal transduction cascades. Previous work from our laboratory showed that stimulation of quiescent fibroblasts with α-thrombin leads to the production of two lipid second messengers in the nucleus: an increase in nuclear diacylglycerol mass and an activation of phospholipase D, which catalyzes the hydrolysis of phosphatidylcholine to generate phosphatidic acid. Diacylglycerol kinase (DGK) catalyzes the conversion of diacylglycerol to phosphatidic acid, making it an attractive candidate for a signal transduction component. There is substantial evidence that this activity is indeed regulated in a number of signaling cascades (reviewed by van Blitterswijk, W. J., and Houssa, B. (1999) Chem. Phys. Lipids 98, 95–108). In this report, we show that the addition of α-thrombin to quiescent IIC9 fibroblasts results in an increase in nuclear DGK activity. The examination of nuclei isolated from quiescent IIC9 cells indicates that DGK-θ and DGK-δ are both present. We took advantage of the previous observations that phosphatidylserine inhibits DGK-δ (reviewed by Sakane, F., Imai, S., Kai, M., Wada, I., and Kanoh, H. (1996) J. Biol. Chem. 271, 8394–8401), and constitutively active RhoA inhibits DGK-θ (reviewed by Houssa, B., de Widt, J., Kranenburg, O., Moolenaar, W. H., and van Blitterswijk, W. J. (1999) J. Biol. Chem. 274, 6820–6822) to identify the activity induced by α-thrombin. Constitutively active RhoA inhibited the nuclear stimulated activity, whereas phosphatidylserine did not have an inhibitory effect. In addition, a monoclonal anti-DGK-θ antibody inhibited the α-thrombin-stimulated nuclear activity in vitro. These results demonstrate that DGK-θ is the isoform responsive to α-thrombin stimulation. Western blot and immunofluorescence microscopy analyses showed that α-thrombin induced the translocation of DGK-θ to the nucleus, implicating that this translocation is at least partly responsible for the increased nuclear activity. Taken together, these data are the first to demonstrate an agonist-induced activity of nuclear DGK-θ activity and a nuclear localization of DGK-δ.

Diacylglycerol (DAG) 1 and phosphatidic acid (PA) have several functions in eukaryotic cells. They are intermediates in phospholipid biosynthesis, cofactors for certain membrane-associated activities, and serve as second messengers. In addition to the widely accepted role of DAG in activating protein kinase C, it also modulates other signaling proteins such as the guanine nucleotide exchange factors vav (1) and Ras-guanyl-releasing protein (2). PA is a modulator of signaling components, such as protein kinase C-(3), Ras-GTPase activating protein (4), and phosphatidylinositol 5-kinase (5), and is involved in the regulation of interleukin-2-dependent lymphocyte proliferation (6). Finally, local alterations in membrane physical properties mediated by changes in lipid composition may regulate the function of membrane-binding proteins (recently reviewed in Ref. 7).
Given their critical role in signaling cascades, it is not surprising that the levels of both DAG and PA are tightly regulated. One enzyme reportedly involved in this regulation is diacylglycerol kinase (DGK) (8). DGK catalyzes the phosphorylation of DAG by converting it to PA, thereby reducing one second messenger in exchange for another. Indeed, there is increasing evidence suggesting that DGK is involved in regulating DAG and PA levels in some signaling pathways. For example, DGK has been shown to be modulated in response to various growth factors and hormones, such as noradrenaline (5,9,10), interleukin-2 (6,(11)(12)(13), concanavalin A (14), insulinlike growth factor-I (15), and leptin (16). Additional support for the notion that DGKs are involved in signaling derives from the observations that DGKs have been found in signaling complexes associated with Rac, phosphatidylinositol 5-kinase, Rho-GDI (17), and Ras-GRP (18), and that they are involved in the regulation of cellular growth (6,11,12,15,19). Clearly, understanding the mechanisms that regulate DGK in various signaling cascades will be essential to delineating the physiological roles of this enzyme.
Defining mechanisms involved in regulating DGK activity requires knowledge of the structural features of the enzyme. This analysis is complicated by the fact that nine mammalian DGK isoforms, organized into five subfamilies, have been identified. All mammalian DGKs contain a conserved catalytic region and at least two cysteine-rich domains. However, they are distinguished by the presence of specific domains, such as EF hands (DGK-␣, DGK-␤, and DGK-␥), pleckstin homology (DGK-␦, DGK-, and DGK-), ephrin homology (DGK-␦), myristoylated alanine-rich C kinase substrate homology domain (DGK-and DGK-), or ankyrin repeats (DGK-␦ and DGK-) (recently reviewed in Refs. 20 and21). The structural diversity among the DGK family members suggests that each isoform performs specific functions and is regulated by different mechanisms. This hypothesis is underscored by observations that DGK isoforms show distinct tissue-specific and subcellular distributions (20,21). Interestingly, a nuclear localization has been reported for four DGK isoforms, DGK-␣, DGK-, DGK-, and DGK-. Wada et al. (14) show that DGK-␣ translocates to the nuclear matrix in rat thymocytes and in peripheral T-lymphocytes. DGK-overexpressed in COS-7 cells was found in the nucleus (22). Interestingly, the data in this study indicates that a protein kinase C-mediated phosphorylation resulted in the redistribution of the enzyme to the cytosol (22). More recently, it was shown that noradrenaline induces a redistribution of DGK-concomitant with an increase in DGK-activity in rat small arteries (10).
Our laboratory and others have focused on the study of agonist-induced nuclear lipid-metabolizing enzymes in IIC9 fibroblasts. Northwood et al. (23) have shown that serum stimulation of quiescent IIC9 cells causes an increase of phosphatidylcholine (PC) biosynthesis associated with a redistribution of CTP:phosphocholine cytidylyltransferase from the nucleus to the endoplasmic reticulum after 10 min. We have demonstrated that the stimulation of quiescent IIC9 cells with ␣-thrombin leads to an increase in nuclear PC-derived DAG mass and an activation of nuclear phospholipase D (24,25). In the studies presented in this report, we demonstrate that another nuclear lipid-metabolizing enzyme, DGK-, is modulated in response to ␣-thrombin in IIC9 cells. These cells express three DGK isoforms, DGK-, DGK-␦, and DGK-, that are highly homologous to the human isoforms. Two of these isoforms, DGK-␦ and DGK-, are present in the nucleus. The addition of ␣-thrombin to quiescent IIC9 cells results in the translocation of DGK-to the nucleus and an increase in nuclear DGK-activity.

EXPERIMENTAL PROCEDURES
Materials-Tissue culture media components and Geneticin (G418) were purchased from Roche Molecular Biochemicals. Plastic culture dishes were purchased from Falcon Labware. Highly purified human thrombin (ϳ4000 NIH units/ml) and bovine serum albumin (radioimmunoassay grade, fraction V) were purchased from Sigma. Cytoscint scintillation-counting fluid was obtained from ICN. [␥-32 P]ATP was purchased by PerkinElmer Life Sciences. Molecular biology enzymes were purchased from Stratagene, Life Technologies, Inc., New England Biolabs, and Roche Molecular Biochemicals. Silica Gel 60 TLC plates (aluminum sheets) were purchased from EM Science. Novafector was purchased from Venn Nova. Anti-DGK-␣, DGK-␦, and DGK-␥ antibodies were kindly provided by Dr. F. Sakane (Department of Biochemistry, Sapporo Medical University School of Medicine, Japan). Anti-DGK-⑀ and DGK-antibodies were kindly provided by Dr. M. Topham (The Huntsman Cancer Institute, Department of Internal Medicine, University of Utah, Salt Lake City, UT). Anti-DGK-monoclonal antibody was purchased from Transduction Laboratories (Lexington, KY). Anti-tubulin antibody was purchased from Zymed Laboratories, Inc. Constitutively active V14RhoA was kindly provided by Dr. J. Exton (Howard Hughes Medical Institute and Department of Molecular Physiology and Biophysics, Vanderbilt University School of Medicine, Nashville, TN). Pig DGK-␣, rat DGK-␤, hDGK-␥, and hDGK-␦ inserted in pSRE were kindly provided by Dr. F. Sakane. hDGK-cDNA inserted in pcDNA3 vector was kindly provided by Dr. W. J. van Blitterswijk (Division of Cellular Biochemistry, The Netherlands Cancer Institute, Amsterdam). Fluorescein-conjugated anti-rabbit was purchased form Amersham Pharmacia Biotech. Texas Red-conjugated anti-mouse IgG was from Jackson ImmunoResearch Laboratories, Inc. (West Grove, PA). Other chemicals were of reagent grade.
Cells and Cell Cultures-IIC9 cells, a subclone of Chinese hamster embryo fibroblasts, were grown, maintained, and serum starved as described previously (26). The cultures were grown and maintained in ␣-minimal essential medium/Ham's F-12 medium (1:1, v/v) containing 5% (v/v) fetal calf serum, 100 units/ml of penicillin, 100 mg/ml strepto-mycin, and 2 mM L-glutamine (complete medium). Subconfluent (60 -70%) cultures were serum-deprived by washing with Dulbecco's modified Eagle's medium and incubated in fresh Dulbecco's modified Eagle's medium containing 100 units/ml of penicillin, 100 mg/ml streptomycin, 2 mM L-glutamine, and 20 mM Hepes for 48 h. After 48 h of serum deprivation, cultures were incubated in fresh Dulbecco's modified Eagle's medium with or without 2 NIH units/ml of ␣-thrombin for the indicated times. Incubations were terminated by removing ␣-thrombincontaining medium and washing cultures with ice-cold PBS. Cells were harvested in ice-cold PBS.
Transient Transfections and Constructs-Transient transfections of IIC9 cells were performed with Novafector (Venn Nova) following the manufacturer's protocol. Cells were grown in 150-cm 2 flasks until 50 -60% confluent. Cultures were washed with PBS and incubated with 10 ml of OPTIMEM containing 20 g of the cDNA-containing vector and 100 l of Novafector. After an overnight incubation, 10 ml of 10% fetal calf serum-containing complete medium was added, and cells were grown for an additional day before harvesting. Transfection efficiency was quantified in every experiment by cotransfection with pEGFP-N3 vector (CLONTECH).
Preparation of Total Homogenates, Cytoplasmic Fractions, and Nuclear Samples-Total homogenates, cytoplasmic fraction, and nuclear samples were prepared as described previously (27) with minor modifications. Cultures were harvested in ice-cold PBS. Cells containing suspensions were centrifuged at 1000 rpm in tabletop Sorval centrifuge for 10 min. Total homogenates were prepared by resuspension of the cellular pellets in Fractionation Buffer A (10 mM Tris, pH 7.5, 10 mM NaCl, 1 mM EDTA and 0.5 mM EGTA, 1 mM phenylmethylsulfonyl fluoride, 10 g/ml aprotinin, 10 M leupeptin, 200 M 2-nitro-4-carboxyphenyl N,N-diphenyl carbonate, 20 M quinacrine, 20 mM NaF) and homogenization on ice in a 2-ml Dounce homogenizer with a type A pestle. Nuclei were isolated from the total homogenates by centrifugation at 400 g (4°C) for 15 min. The resulting supernatant, i.e. cytoplasmic fraction, was removed, and the nuclei were purified as follows: the nuclear pellet was resuspended in 500 l of Fractionation Buffer A and homogenized on ice in a 2-ml Dounce homogenizer with a type A pestle. The nuclei-containing suspension was layered on top of 1 ml of 45% sucrose cushion in a 1.6-ml microcentrifuge tube and centrifuged at 10,000 g (4°C) for 20 min. The purified nuclei were either resuspended in the appropriate buffer for each experiment or disrupted as described previously (28). The purified nuclei were resuspended in 5 mM Tris-HCl, pH 8.0, 0.1 mM EDTA, 100 units/ml of DNase I and allowed to swell by incubating on ice for 10 min preceding disruption by 60 passages through a 25-gauge hypodermic needle. The protein concentration of all samples was assayed using BCA Protein Assay Kit (Pierce). The purity of nuclear preparations was assessed by immunoblotting for tubulin.
DGK Enzymatic Assays-DGK activity was assayed in vitro using octylglucoside (OG)/DAG mixed micelles. OG/DAG mixed micelles were prepared as follows: a mixture of 0.25 mM dioleoylglycerol, 55 mM OG, and PS (either 1 mM resulting in 1.8 mol% in micelles or 5 mM resulting in 8.3 mol% in micelles) was resuspended in 1 mM diethylenetriamine pentaacetic acid, pH 7.4, by vortexing and sonicating until the suspension appeared clear. 20 l of mixed micelles were added to 70 l of the reaction mix (final concentration: 100 M diethylenetriamine pentaacetic acid, pH 7.4, 50 mM imidazole-HCl, 50 mM NaCl, 12.5 mM MgCl 2 , 1 mM EGTA, 1 mM dithiothreitol, 1 mM [␥-32 P]ATP). 10 l of total homogenate or isolated nuclei were added to 90 l of mixed micelles reaction mix solution. The reaction was started by vortexing for 3 s and sonicating for 5 s. After a 30-min incubation at 25°C, the reaction was terminated by the addition of chloroform/methanol/1% perchloric acid (1:2: 0.75, v/v) and then vortexed. 1% perchloric acid and chloroform (1:1, v/v) were added, and the mixture was centrifuged for 5 min at 2000 rpm in a tabletop Sorval centrifuge at 25°C. The organic phase was washed twice in 1% perchloric acid, an aliquot was dried under a stream of nitrogen and spotted on a Silica Gel 60 TLC plate. PA was separated by developing in chloroform:acetone:methanol:acetic acid: water (10:4:3:2:1) (v:v). The amount of [␥-32 P]PA was measured by liquid scintillation spectrophotometer. DGK activity was linear with respect to time (15-60 min) and protein concentration (8 -150 g).
Western Blot Analysis-Total homogenates or nuclear samples were resuspended in Laemmli sample buffer. 20 g of protein were separated by SDS-PAGE in an 8% acrylamide gel. Proteins were transferred to a polyvinylidene difluoride membrane (Bio-Rad) and blocked for 1 h in TBS (TBS, pH 7.4, containing 0.1% (v/v) Tween 20) supplemented with 5% nonfat dry milk. The membrane was incubated with the indicated antibodies in TBS for 2 h, and the immunoreactive bands were visualized using peroxidase-conjugated anti-rabbit or anti-mouse IgG anti-bodies (Sigma) and enhanced with chemiluminescence (ECL kit, Amersham Pharmacia Biotech).
Immunofluorescence Microscopy of DGK-and DGK-␦-IIC9 cells were grown on glass coverslips, serum-deprived, and ␣-thrombin-stimulated in the conditions described previously. After ␣-thrombin stimulation, the medium was aspirated, and cells were washed twice with ice-cold PBS. After fixation in methanol:acetone (1:1, v/v) for 20 min at Ϫ20°C, cells were washed with PBS and rehydrated by incubation in TBS for 10 min at room temperature. Fixed cells were then incubated with primary antibody (monoclonal anti-DGK-(1:10), polyclonal anti-DGK-␦ (1:100), or monoclonal anti-tubulin (1:100)) for 1 h. After three washes in TBS, cells were incubated with either Texas Red-conjugated anti-mouse IgG or fluorescein-linked anti-rabbit secondary antibody for 45 min. Cells were then washed in TBS, and nuclei were stained by incubating with Hoechst stain for 3 min. Finally, cells were washed in distilled water, and the coverslips were mounted on slides on a drop of glycerol containing 0.1 M N-propyl gallate. Images were acquired using a microscope (Axioskop, Zeiss, Inc., Thornwood, NY) equipped with epifluorescence and a charge coupled device camera (Photometric Sensys, Tucson, AZ) using IP Lab software (Signal Analytic Corp., Vienna, VA).

RESULTS
Basal DGK Activity in IIC9 Nuclei: Inhibition by PS-Interest in eukaryotic DGKs and their subcellular localization is increasing. Given recent reports of DGKs being present in the nucleus (8,10,14,21,22) and our interest in nuclear lipid metabolism (29), we determined whether nuclei isolated from quiescent IIC9 fibroblasts possessed a DGK activity. The presence of this activity in nuclei isolated from IIC9 cells was quantified in vitro using OG/DAG mixed micelles as described under "Experimental Procedures." Because PS is an activator for most DGKs, the nuclear DGK activity was assayed in the presence of either low (1.8 mol%) or high (8.3 mol%) concentrations of PS. Interestingly, increasing the PS concentration from 1.8 mol% to 8.3 mol% resulted in a 3-fold decrease in DGK activity quantified in these nuclei (from 3 ϫ 10 Ϫ5 nmol/g/min to 1 ϫ 10 Ϫ5 nmol/g/min) (Fig. 1).
DGK-␦ and DGK-Are Present in IIC9 Nuclei-The above data showed that the nuclear DGK activity in quiescent IIC9 cells was inhibited by a higher concentration of PS (8.3 mol%), suggesting that a PS-sensitive DGK isoform was responsible for the activity. Because only the mammalian DGK-␦ isoform is known to be inhibited by PS (30), it was important to determine whether the DGK-␦ isoform was present in the isolated nuclei. Western blot analysis of isolated nuclei was performed using antibodies raised against human DGK isoforms ␣, ␥, ␦, ⑀, , and . This analysis indicated that two DGK isoforms, DGK-␦ and DGK-, were present in IIC9 nuclei (Fig. 2, A and B). In addition to these isoforms, Western blot analysis of total cell homogenates showed that IIC9 cells also express DGK-, but this isoform was present only in the cytoplasmic fraction of quiescent IIC9 cells (Fig. 2C). We further analyzed the endogenous mRNAs using a PCR-based amplification of a cDNA phage library of IIC9 cells. Oligonucleotide primers targeted to conserved regions of 10 cloned DGKs were generated and used in PCR-based reactions (see under "Experimental Procedures"). The 600-base pair PCR product was sequenced and aligned to all the non-redundant GenBank TM CDS translations, Protein Data Bank, Swiss Protein Database, Protein Identification Resource, and PRF nucleotide sequences using the BLASTN program. This analysis revealed the presence of an mRNA with high homology to the human DGK-cDNA. A second set of PCR reactions was performed using primers homologous to very specific regions of DGK isoforms -␣, -␤, -, -, and -(see under "Experimental Procedures"). Only the reaction performed with the specific primers for DGK-gave a PCR product from which the sequence was highly homologous to human DGK-(data not shown).
␣-Thrombin Stimulates a Nuclear DGK Activity That Is Not Inhibited by PS-In previous studies, our laboratory demonstrated that ␣-thrombin stimulation of IIC9 cells results in the activation of a nuclear PC-PLD and an increase in nuclear DAG mass (24,27). In view of our observation that IIC9 nuclei contain a DGK activity, we asked whether nuclear DGK activity was also modulated in response to ␣-thrombin. IIC9 cells were brought to quiescence by serum deprivation for 48 h. Quiescent cells were stimulated with ␣-thrombin (2 NIH units/ ml) for 0, 1, 3, 5, 10, 15, and 30 min at 37°C. Nuclei were isolated, and DGK activity was quantified using OG/DAG mixed micelles containing either no PS or 8.3mol% PS. As shown in Fig. 3A, nuclear DGK activity peaked within 3-5 min and then declined to a level near basal at 30 min. Consistent with results shown in Fig. 1, the basal nuclear DGK activity decreased when assayed in the presence of high PS (8.3 mol%) (Fig. 3B). However, the relative increase of ␣-thrombin-induced nuclear DGK activity over the basal nuclear activity was not inhibited by the presence of PS (Fig. 3C). These data indicate that the ␣-thrombin-induced nuclear DGK activity was not inhibited by PS.
DGK-Is Not Inhibited by PS-Taken together, the above data indicate that the basal DGK activity in IIC9 nuclei is inhibited by PS (Figs. 1 and 3, A and B), whereas the ␣-thrombin-stimulated DGK activity is not inhibited by PS (Fig. 3). One explanation for this observation is that DGK-␦, known to be inhibited by PS (30), is responsible for the basal nuclear DGK activity, whereas DGK-is activated upon ␣-thrombin stimulation. To test this hypothesis, we first examined the effect of PS on DGK-. Although several reports suggest that PS activates DGK-, there are no published data demonstrating the effect of PS on DGKactivity. We assayed the DGK activity in total cellular homogenates prepared from cells overexpressing DGK-and from wild-type cells in the presence of both high (8.3 mol%) and low (1.8 mol%) concentrations of PS. As expected, DGK activity in total homogenate from IIC9 cells over-expressing DGK- (Fig. 4, right bars) was greater than the activity in wild-type cells (Fig. 4, left bars). Interestingly, the DGK activity in total homogenates from cells overexpressing DGK-was increased with increasing concentration of PS from 1.8 mol% (Fig. 3, black right bar) to 8.3 mol% (Fig. 4, gray right  bar). The above data indicate that DGK-is not inhibited by PS.
␣-Thrombin-stimulated Nuclear DGK Activity Is Inhibited by Constitutively Active RhoA-Because DGK-is not inhibited by PS (Fig. 4) while DGK-␦ is inhibited by this lipid (30), the above data suggest that the ␣-thrombin-stimulated nuclear DGK activity is due to an increase in nuclear DGK-activity. To further test this possibility, we took advantage of an interesting result published by Houssa et al. (31) that shows that constitutively active RhoA binds and inhibits DGK-. To determine whether the constitutively active mutant of RhoA, V14RhoA, could be used as a tool to discriminate between DGK-and DGK-␦ activity, we initially determined whether V14RhoA binds to DGK-␦. Cell lysates from IIC9 cells overexpressing either DGK-␦ or DGK-were combined with cell lysates overexpressing V14RhoA. After incubation for 15 min at 4°C, V14RhoA was immunoprecipitated, and the immunoprecipitates were subjected to Western blot analysis with either anti-DGK-␦ (Fig. 5A) or anti-DGK- (Fig. 5B) antibodies. The result showed that DGK-, but not DGK-␦, coimmunoprecipitated with V14RhoA.
The above results indicated that V14RhoA can be used to discriminate between activation of DGK-␦ and DGK-. Therefore, the effect of overexpressing V14RhoA on ␣-thrombin-induced nuclear DGK activity was analyzed. Nuclei were isolated from V14RhoA-expressing, quiescent or ␣-thrombin-stimulated cells, and nuclear DGK activity was quantified. As shown in Fig. 6, V14RhoA does not affect the basal DGK activity but eliminates the ␣-thrombin-induced increase in nuclear DGK activity. These data indicated that nuclear DGK-is activated in response to ␣-thrombin. We should note that the overexpression of a dominant negative RhoA mutant (RhoA19) had no effect on the nuclear peak of DGK activity observed at 3-5 min after ␣-thrombin stimulation (data not shown).
Anti-DGK-Antibody Blocks the ␣-Thrombin-stimulated Nuclear DGK Activity in Vitro-To further examine the hypothesis that the nuclear DGK-is activated by ␣-thrombin stimulation, we measured the DGK activity in nuclei incubated with the anti-DGK-monoclonal antibody. This antibody binds to DGK-in the catalytic domain (amino acids 677-883) and inhibits its activity (data not shown). Nuclei isolated from quiescent cells or cells stimulated with ␣-thrombin were disrupted and incubated overnight (4°C) with constant agitation in the presence or absence of the monoclonal anti-DGK-antibody. DGK activity was then measured. As shown in Fig. 7, the presence of anti-DGK-monoclonal antibody dramatically inhibited the ␣-thrombin-stimulated nuclear DGK activity. As a control to verify that the inhibitory effect was not a general one due to the presence of any antibody, we performed the same experiment with an anti-DGK-␣ antibody. As shown in Fig. 7, the ␣-thrombin-induced nuclear DGK activity was inhibited only in the presence of anti-DGK-antibody.
DGK-Translocates to the Nucleus-The above data demonstrate that the nuclear DGK-activity is increased upon ␣-thrombin stimulation. The mechanism responsible for this increased activity is clearly one of the next important questions. Studies have shown that agonist stimulation results in a redistribution of some DGK isoforms. For example, DGK-␣ translocates to the nuclear matrix when rat thymocytes and peripheral T-lymphocytes are stimulated with concanavalin A or anti-T-cell receptor antibody (14). It is possible, therefore, that at least part of the ␣-thrombin-induced increase in nuclear DGK-activity results from a translocation of the enzyme to the nucleus.
To investigate this possibility, the amount of endogenous DGK-present in the nuclei isolated from quiescent and stimulated IIC9 cells was examined by Western blot analysis. The immunoblot in Fig. 8A shows that the level of nuclear DGKincreased within 3-5 min following ␣-thrombin stimulation, remained elevated for 10 min, and then declined toward the basal level within 30 min. In contrast, there was no change in the amount of DGK-␦ or DGK-at the nucleus after 5 min of ␣-thrombin stimulation (Fig. 8, B and C).
To further examine the DGK-translocation to the nucleus, we performed indirect immunofluorescence microscopy on intact cells. IIC9 cells were grown on glass coverslips and serumdeprived. The quiescent IIC9 cells were incubated with ␣-thrombin for 0 min (Fig. 9, G and I) or 5 min (Fig. 9, H and J) and then fixed and permeabilized as described under "Experimental Procedures." The cells were first incubated with the anti-DGK-monoclonal antibody (Fig. 9, G and H) or anti-DGK-␦ polyclonal antibody (Fig. 9, I and J) and then incubated with a Texas-Red-conjugated anti-mouse antibody or fluorescein-linked anti-rabbit Ig. To verify that the nuclear labeling was not attributed to a nonspecific binding of the secondary antibodies, cells were incubated with the Texas-Red-conjugated anti-mouse antibody alone (Fig. 9, A and B), with the fluorescein-linked anti-rabbit antibody alone (Fig. 9, C and D), or with monoclonal anti-tubulin antibody followed by Texas-Red-conjugated anti-mouse antibody (Fig. 9, E and F). Fig. 9, G and H, shows that DGK-translocates to the nucleus after 5 min of ␣-thrombin stimulation. Consistent with the Western blots analysis (Fig. 8B), the amount of nuclear DGK-␦ does not change (Fig. 9, I and J). The above data indicate that the translocation of DGK-is at least partly responsible for the ␣-thrombin-induced increase in nuclear DGK-activity. DISCUSSION DGK catalyzes the transformation of one lipid second messenger, DAG, into another messenger, PA, and its activity is Quiescent IIC9 cells were stimulated with ␣-thrombin for 0 or 5 min. Nuclei were isolated and disrupted, and each nuclear sample was divided into two aliquots. Either monoclonal antibody anti-DGK-(50 ng/l) (right bar) or polyclonal anti-DGK-␣ antibody (1:50) (middle bar) was added to the first aliquot, and both aliquots were incubated overnight at 4°C with constant agitation. DGK activity was measured with the mixed micelles assay containing octylglucoside (55 mM), DAG (0.4 mol%), and PS (8.3 mol%). All presented data are from two independent experiments each performed in duplicate, except for the data marked with an asterisk that are from one experiment performed in duplicate. modulated in response to various growth factors and hormones, such as noradrenaline (9), interleukin-2 (11,12,14), insulinlike growth factor-I (15), and leptin (16). Defining the precise mechanisms of regulation and the physiological role of DGKs in any signaling cascade requires knowledge of (a) the particular DGK isoform involved and its molecular characteristics, (b) cellular localization/translocation of the DGK, and (c) the interaction of the enzyme with other signaling proteins. In the present study, we analyze the nuclear DGK activity in IIC9 fibroblasts and its response to ␣-thrombin stimulation.
Since the first observation was made that nuclei contain lipid-metabolizing activities, which are modulated in response to agonists (24,(33)(34)(35)(36)(37)(38), a considerable effort has been devoted to identifying the physiological roles and regulatory mechanisms of these enzymes in the nucleus. The evidence for a nuclear DGK activity was first demonstrated in 1983 by Smith and Wells (39). More recently, the nuclear DGK activity has been characterized in rat liver nuclei (40,41), and some agonist-modulated DGK isoforms, DGK-␣, DGK-, and DGK-, have been localized in the nucleus of a variety of cells (14). DGK-has also been localized at the nucleus in rat mesenteric small arteries (10), as well as N115 neuroblastoma cells, 2 MelJuso cells and COS-7 cells. 3 In this report, we show that ␣-thrombin stimulates a nuclear DGK activity in IIC9 fibroblasts (Fig. 3, A and B). In these cells, three DGK isoforms are present: DGK-␦, DGK-and DGK- (Fig.  2). In quiescent and ␣-thrombin-stimulated IIC9 cells, only the DGK-␦ and DGK-isoforms are present at the nuclear level (Fig.  2). To determine whether the ␣-thrombin-stimulated activity was 2   due to DGK-␦ or DGK-activation or both, we took advantage of the observations that DGK-␦ is inhibited by high concentrations of PS (30) while DGK-is inhibited by constitutively active RhoA mutant V14RhoA (31). The observation that the ␣-thrombininduced increase in nuclear DGK activity was not inhibited by high concentrations of PS but inhibited by the expression of V14RhoA indicates that the nuclear DGK activity being modulated is DGK-. Furthermore, the incubation of disrupted nuclei with the anti-DGK monoclonal antibody strongly decreased the stimulated nuclear activity (Fig. 7) but did not affect the basal activity (data not shown). In the same experiment, the addition of anti-DGK-␣ antibody instead of anti-DGK-antibody had no affect on the stimulated or unstimulated nuclear DGK activity. This result further demonstrates that nuclear DGK-is activated by ␣-thrombin stimulation.
The mechanisms by which agonist-induced DGK activity is regulated have not been clearly defined. In this regard, it is interesting to note that DGK activity has been shown to shuttle between different cellular compartments in response to agonist stimulation (10,14,32,(42)(43)(44)(45)(46)(47)(48)(49). Recent observations have shown that DGK-is activated and redistributed among intracellular compartments upon noradrenaline stimulation of rat small arteries (10). In this report, we show that in IIC9 cells, the ␣-thrombin-induced increase in nuclear DGK-activity is accompanied by a translocation of the enzyme to the nucleus (Figs. 8 and 9). These data provide compelling evidence for an induced nuclear translocation of the enzyme being at least partly responsible for the induced increase in nuclear DGK-activity.
Our data indicate that the ␣-thrombin-induced increase in nuclear DGK-activity peaks within 3-5 min of stimulation and then begins to decline toward basal levels. It is possible that RhoA plays a role in this inhibition, because constitutively active RhoA (V14 RhoA) binds to DGK-and inhibits its activity (31). In a previous report, we presented evidence showing that RhoA translocates to the nucleus of IIC9 cells within 10 min following ␣-thrombin stimulation (24). Therefore, it is reasonable to suggest that in addition to activating nuclear PC-PLD (24), the ␣-thrombin-induced translocation of RhoA to the nucleus also plays a role in modulating DGK-activity.
Taken together, these results potentially add an interesting dimension to the regulation of nuclear PA production. There is a reciprocal relationship between the kinetics of nuclear DGKand PC-PLD activation. Nuclear DGK-activity is increased at the nucleus after 3-5 min of ␣-thrombin stimulation and begins to decline after approximately 10 min. Nuclear PC-PLD activity, on the other hand, peaks after approximately 10 min of ␣-thrombin stimulation because of the translocation of RhoA to the nucleus and is elevated for at least 1 h (24). These data suggest the following hypothesis (Fig. 10). The addition of ␣-thrombin to quiescent IIC9 cells leads to the translocation of DGK-to the nucleus within 3-5 min, resulting in an increase in DGK--mediated PA production. Subsequent to this activation, RhoA translocates to the nucleus where it inhibits DGKand activates PC-PLD (24). In this manner, RhoA acts to mediate a "switch" from DGK--mediated PA production to PC-PLD-mediated PA production. The reason for this switch is not clear, but it may be involved in generating PA in distinct subnuclear localizations and/or different molecular species of PA. We should note that nuclear translocation alone may not be sufficient for nuclear DGK-activation. For example, this activation may require the production of a cofactor, such as phosphoinositides phosphorylated by a phosphatidylinositol 3-kinase (10). Studies are underway to test this hypothesis and examine these possibilities.