Nuclear ADP-ribosylation factor (ARF)- and oleate-dependent phospholipase D (PLD) in rat liver cells. Increases of ARF-dependent PLD activity in regenerating liver cells.

Two forms of phospholipase D (PLD) have been found to be present in nuclei isolated from rat hepatocytes by measuring phosphatidylbutanol produced from exogenous radiolabeled phosphatidylcholine in the presence of butanol. In nuclear lysates from either rat liver or ascites hepatoma AH 7974 cells, the PLD activity was markedly stimulated by a recombinant ADP-ribosylation factor (rARF) in the presence of the guanosine 5′-O-(3-thiotriphosphate) (GTPγS) and phosphatidylinositol 4,5-bisphosphate. ATP and phorbol-12-myristate 13-acetate had no synergistic effect on this PLD activity. On the other hand, the nuclear PLD was stimulated by unsaturated fatty acids, especially by oleic acid. The ARF-dependent nuclear PLD activity was increased in the S-phase of the regenerating rat liver after partial hepatectomy and also was much higher in AH 7974 cells than in the resting rat liver. In contrast, the levels of the oleate-dependent PLD activity remained constant throughout the cell cycle in liver regeneration. The intranuclear levels of the stimulating proteins of the nuclear PLD activity, e.g. ARF, RhoA, and protein kinase Cδ increased in the S-phase of the regenerating liver. These results suggested that the nuclear ARF-dependent PLD activity may be associated with cell proliferation.

It has been known that cell nuclei contain a variety of enzymes generating lipid second messengers, such as sphingomyelinase (1), phospholipase A2 (2), PI 1 -specific phospholipase C (3)(4)(5)(6) and lipid kinases (7). Growth factors seem to be able to affect the phosphoinositide metabolism in nuclei, suggesting a subtle regulation during cell activation (8,9). In addition, accumulation of diacylglycerol (DG) in nuclei and translocation of protein kinase C to them have been demonstrated in a variety of cell types (10,11). A large rise in mass of DG, with only small changes in mass of phosphoinositides, suggested a source of DG other than polyphosphoinositides, for example, phosphatidylcholine (PC), in nuclei (12). A recent study (13) has demonstrated that phosphatidylethanol formation from phosphatidylcholine, which was specifically catalyzed by phospholipase D (PLD) in the presence of ethanol, was induced by PMA in nuclei isolated from kidney cells and that nuclei possess the ability to generate DG and phosphatidic acid through the PLD:phosphatidic acid phosphohydrolase pathway. Thus, upon cell stimulation with agonists, the enzymes involving PC metabolism are considered to be activated in the nucleus as well as in the plasma membrane.
It was demonstrated that PLD activity can be regulated by several factors (14,15); oleate, small G proteins (ARF and Rho family), phosphatidylinositol 4,5-bisphosphate (PIP 2 ), phosphatidylethanolamine (PE), Ca 2ϩ , protein kinase C, protein tyrosine kinase. Recently, two different types of PLD, oleatedependent and ARF-dependent, were isolated from rat brain membranes (16). The oleate-dependent PLD was purified from pig lung microsomes (17), and ARF-dependent PLD activity was found to be abundant in Golgi-enriched membranes from several cell lines (18). More recently, a gene of human ARF-dependent PLD (hPLD1) has been cloned (19). It has been shown that the membrane-associated PLD activity was modulated by protein kinase C and Rho family (20 -24). These observations suggest the presence of different isoforms of PLD.
The nuclear PLD activity from Madin-Darby canin kidney (MDCK)-D1 cells was observed to be regulated by protein kinase C and RhoA (23). It was also shown that an oleate-dependent form of PLD was present in rat brain neuronal nuclei and inhibited by acidic phospholipids, such as PIP 2 (25). However, the physiological roles of PLD in nuclei remain unclear.
In this study, we have demonstrated that two forms, ARFdependent and oleate-dependent PLD activities, were present in the isolated nuclei from rat resting liver, regenerating liver, and rat ascites hepatoma cells AH7974 and characterized their PLD activities using exogenous radiolabeled PC substrates. Furthermore, we have shown a transient elevation of the nuclear ARF-dependent PLD activity, but not the oleate-dependent PLD activity during S-phase of liver regeneration.
Isolation of Nuclei-Nuclei were isolated and purified from rat ascites hepatoma AH7974 cells and rat liver cells as described previously (2,26). They were highly pure as examined by electron microscopic examination and by determining marker enzymes for microsomes (glucose-6-phosphatase), plasma membrane (5Ј-nucleotidase), and mitochondoria (cytochrome oxidase). DNA polymerase ␣ activity was measured in isolated nuclei as described previously (27). The isolated nuclei were resuspended in 0.25 M sucrose, 10 mM MgCl 2 , 10 mM Tris-HCl, pH 7.4, and 1 mM phenylmethylsulfonyl fluoride.
Assay of Phospholipase D Activity-The nuclear fractions were precipitated by centrifugation at 10,000 ϫ g for 5 min and resuspended in lysis buffer containing 25 mM HEPES/NaOH buffer, pH 7.4, 100 mM KCl, 5 mM MgCl 2 , 1 mM phenylmethylsulfonyl fluoride, and 10 mg/ml leupeptin. The suspended nuclei were sonicated three times 15 s each. The PLD activities were assayed by slight modification of the procedure of Massenburg et al. (16). For the assay of oleate-dependent PLD activity, egg PC vesicles (20 l) containing 15 nmol of egg PC and 400,000 dpm of [ 3 H]DPPC were prepared by sonication and then mixed with 5 l of 12 mM sodium oleate. The substrates were added to nuclear lysates in a total volume of 120 l containing 50 mM HEPES/NaOH buffer, pH 7.0, 2 mM EGTA, 500 mM KCl, 1 mM MgCl 2 , 2.1 mM CaCl 2 , and 0.3% (v/v) butanol and then incubated at 30°C for 30 min.
Reactions were terminated by addition of 0.8 ml of chloroform/methanol (1:2, v/v) and 0.4 ml of chloroform and 0.3 ml of 200 mM KCl, 5 mM EDTA. Lipids were extracted by the method of Bligh and Dyer (29) and separated on TLC plates (Silica gel 60A) in a solvent system using the upper phase of ethyl acetate/2,2,4-trimethylpentane/acetic acid/water (13:2:3:10, v/v) as described by Chalifa et al. (30). The plates were exposed to I 2 vapor and [ 3 H]PBut formation was identified by comigration with the authentic standard. Silica gels scraped off from the plates were mixed with scintillation mixture, and radioactivities were counted in a liquid scintillation counter.
Gel Electrophoresis and Immunoblotting-Protein concentration was determined using a Bio-Rad Bradford assay reagent using bovine serum albumin as standard. The isolated nuclei were resuspended in the sample buffer (60 mM Tris, 10% glycerol, 45 mM mercaptoethanol, 80 mM sodium dodecyl sulfate, pH 6.8) and boiled for 3 min. Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis using 13 and 8% polyacrylamide gels and transferred to the polyvinylidene difluoride membrane and then incubated with antibodies. Antigen-antibody complexes were detected by the chemiluminescence (ECL) method (Amersham Corp.). The intensities of bands were quantitated by densitometry (ATTO Densitograph AE-6900M).
Expression and Purification of Recombinant Proteins-Escherichia coli bearing ARF1 plasmid was gift from Dr. Joel Moss and recombinant ARF1 was purified by chromatography on DEAE-Sephacel and Sephacryl S-300 columns as described (31). Protein preparation showed only one band on Coomassie-stained SDS-polyacrylamide gels.

Characterization of PLD Activity in Rat Liver and Rat Ascites
Hepatoma Cell Nuclei-It has been demonstrated that two types of PLD activity, oleate-dependent and ARF-dependent, have been reported in mammalian cells and tissues (14,16,17). PLD activity was examined in nuclei isolated from rat resting liver and AH7974 cells using the exogenous radiolabeled PC substrates in the presence or absence of rARF (5 M) and GTP␥S (30 M) or oleate (0.5 mM). As shown in Table I, the specific activities of the ARF-dependent and oleate-dependent PLD of AH7974 nuclei were approximately 10-and 3-fold higher than those of rat resting liver nuclei, respectively. When used PE/PIP 2 /PC vesicles as substrate, the PLD activity without stimulators was high in the AH7974 nuclei. In the AH7974 nuclei, as shown in Fig. 1, the [ 3 H]PBut formation in the absence of GTP␥S showed a linear increase in a time-dependent manner, although a much less increase than that with GTP␥S, indicating the presence of a GTP␥S-independent PLD activity. Addition of GTP␥S (30 M) to AH7974 nuclear lysate increased the PLD activity in a time-dependent manner (Fig.  1A). The rate of [ 3 H]PBut formation was linear for at least 60 min. The PLD activity without GTP␥S was not affected by ATP (0.5 mM). The PLD activities with or without GTP␥S were increased by PIP 2 in a dose-dependent manner (Fig. 1B). The ratio of PIP 2 :PC was 3:1 for the maximal activation. The PLD activity without GTP␥S in the nuclei isolated from rat resting liver was very low, even in the presence of PIP 2 .
The nuclear PLD activities of both rat resting liver and AH7974 cells were increased by rARF in the presence of 30 M GTP␥S in a dose-dependent manner, reaching a maximal level at 8 M ( Fig. 2A). Linear increases in the ARF-dependent PLD activity were observed up to 2 and 5 g of proteins of the AH7974 and rat resting liver nuclear lysates, respectively (Fig.  2B). Both nuclear ARF-dependent PLD activities were limitedly stimulated by low concentration of Ca 2ϩ with maximal effect at 0.3 M and inhibited by higher concentrations than 1 M (Fig. 2C). Mg 2ϩ stimulated both nuclear ARF-dependent PLD activities with maximal concentration of 3 mM (Fig. 2D).
It has been known that PKC had a synergistic stimulatory effect of the ARF-dependent PLD activity (32). To examine the effect of PKC on the nuclear ARF-dependent PLD activity, the nuclear lysates of the rat resting liver and AH7974 cells were incubated with GTP␥S (30 M), rARF (5 M), PMA (100 nM), and ATP (0.5 mM) or their combination. As shown in Fig. 3, rARF was most potent activator for both nuclear PLD activities. PMA had no synergistic stimulatory effect on the nuclear ARF-dependent PLD activity in both nuclei.
To examine the oleate-dependent PLD activity, the rat liver and AH7974 nuclear were incubated with [ 3 H]DPPC in egg PC vesicles as substrate in the presence of 0.5 mM oleate. The rate of [ 3 H]PBut formation of their nuclear lysates was linear for at least 30 min, and a linear increase of [ 3 H]PBut formation was observed up to 50 g protein of their nuclear lysates (data not shown). Characterization of the oleate-dependent PLD activity of the AH7974 nuclei was shown in Fig. 4. An optimal pH was 6.0 -7.0 in 50 mM HEPES/NaOH buffer (Fig. 4A). Lower concentration of Ca 2ϩ less than 0.1 mM had little effect on the [ 3 H]PBut formation induced by oleate, but its higher concentrations over 0.5 mM were rather inhibitory. On the other hand, Mg 2ϩ stimulated the PLD activity with the maximal level at 1  (Fig. 4B). Oleic acid was the most potent in stimulating the AH7974 nuclear PLD activity. The [ 3 H]PBut formation was increased by oleate in a dose-dependent manner with a maximal level at 0.5 mM (Fig. 4C). The optimal molar ratio of sodium oleate to egg PC was approximately 4. Other unsaturated fatty acids, linoleic and arachidonic acids, at the same concentration enhanced the PLD activity to lower extents, as compared with oleate (Fig. 4D). In contrast, saturated fatty acids, palmitic and stearic acids, had no effect. The biochemical properties of the oleate-dependent PLD activity in rat liver nuclei were similar to those of AH7974 nuclei (data not shown). The nuclear oleatedependent PLD activity was not affected by GTP␥S, PIP 2 , rARF, or PMA (data not shown), indicating that the nuclear oleate-dependent PLD is distinct from the ARF-dependent one.
Changes of Nuclear PLD Activity in Regenerating Liver Cells-In order to examine whether the nuclear PLD activity is somehow associated with cell proliferation, PLD activity was measured in nuclei isolated from regenerating liver of partially hepatectomized rats. The ARF-dependent PLD activity began to increase 23 h after hepatectomy, reaching a peak at 26 -28 h (Fig. 5B). At 28 h after hepatectomy the nuclear ARF-depend- ent PLD activity was approximately 5-fold higher than that of the control liver nuclei. On the other hand, the oleate-dependent PLD activity remained unchanged during liver regeneration (Fig. 5C). The GTP␥S-independent PLD activity was very low in the regenerating liver nuclei and remained unchanged (Fig. 5A). The transient increase in the ARF-dependent PLD activity was observed 2 ϳ 3 h later following the increase in the DNA polymerase ␣ activity (Fig. 5D), which is a marker for S-phase (3,27), suggesting that the ARF-dependent PLD is induced along progression of DNA synthesis in S-phase (33).
Intranuclear Levels of the ARF-dependent PLD-activating Proteins--It has been reported that cell stimulation with various agonists induces translocation of protein kinases to nuclei as a result of DG formation by PI-PLC activation (10,11). To see the proteins translocated to the nuclei after partial hepatectomy, the nuclei from S-phase liver cells (28 h after partial hepatectomy) were examined by Western blotting with various antibodies. As shown in Fig. 6A, the levels of PKC␣, PKC␦, p42, p38 MAP kinases, and ARF in nuclei isolated from S-phase of the regenerating liver cells were much higher than those of the control nuclei (28 h after sham-operated liver). In contrast, the nuclear PKC␤II level of the regenerating liver cells was lower compared with the control nuclei. Analysis of Western blotting of the AH7974 nuclei revealed the presence of PKC isozymes (␣, ␤I, ␤II, and ␦) (data not shown).
Time-dependent changes in PKC isozyme levels in nuclei during liver regeneration showed different profiles among PKC isozymes (Fig. 6B). There was a remarkable elevation of PKC␦ level in the S-phase cell nuclei. On the other hand, the level of PKC␣ was not changed, and the PKC␤II level rather decreased during liver regeneration. Significant increases in the nuclear levels of p42 and p38 MAP kinases were observed in S-phase.
To examine the intranuclear G protein levels, the isolated nuclei from AH7974, the resting rat liver, and regenerating liver (28 h after partial hepatectomy) were blotted with various specific antibodies. Analysis of the Western blotting revealed the presence of low molecular weight GTP-binding proteins, ARF, RhoA, and Cdc42 and heterotrimeric GTP-binding protein, Gi2 (Fig. 7). The nuclear ARF and RhoA levels of AH7974 cell were 9-and 2-fold higher than those of rat resting liver cells, respectively, although the Cdc42 and Gi2 levels were almost similar between both nuclei. ARF protein was scant in the rat resting liver nuclei, but it was 5-fold increased in the nuclei of S-phase. A little increase of nuclear RhoA level was observed in the S-phase (1.3-fold). DISCUSSION We have demonstrated here that two forms of PLD activity, GTP␥S-dependent and oleate-dependent, were present in the isolated nuclei from rat resting, regenerating liver cells, and rat ascites hepatoma AH7974 cells. Their GTP␥S-dependent PLD activities were markedly stimulated by rARF. The nuclear ARF-dependent PLD activity was very high in the AH7974 cells, and its elevation was observed at S-phase of the regenerating liver cells. The nuclei isolated from the regenerating liver and AH cells contained ARF, RhoA, Cdc42, and various FIG. 5. Changes in the nuclear PLD and DNA polymerase ␣ activities during liver regeneration. Livers were obtained from nontreated, sham-operated rats, and regenerating livers from partially hepatectomized rats (1,2,5,8,12,18,20,22,24,26,28,30, and 32 h post-hepatectomy). Nuclei were prepared from each liver as described under "Experimental Procedures." [ 3 H]PBut formation was measured in nuclear lysates (3 g of protein) of regenerating livers from operated rats (q) and of control livers from sham operated rats (E) by using

FIG. 6. Changes in PKCs and MAP kinases during liver regeneration.
A, the nuclei (50 -200 g of protein) from 28 h liver after partially hepatectomized (ϩ) or sham-operated (Ϫ) rat were separated by SDS-polyacrylamide gel electrophoresis and immunoblotted with various antibodies. A blot representative of three independent experiments is shown. B, the isolated nuclei (50 g of protein) during regenerating liver from partially hepatectomized rats (2,5,12,18,20,24,26,28, and 32 h post-hepatectomy) were separated by SDS-polyacrylamide gel electrophoresis and immunoblotted with various antibodies. Changes of nuclear levels of PKCs and MAP kinases (MAPK) were quantified by scanning densitometry and mass of each PKC and MAP kinase isozyme was designated as 100%. Results are given as mean of three different experiments.
PKC isozymes (PKC␣, PKC␤II, and PKC␦). The ARF-dependent PLD has been known to be activated synergistically by either RhoA, Cdc42, or PKC␣ (21,22,32,34,35). Balboa et al. (13,23) have reported that isolated nuclei from MDCK cells contain an ATP-dependent PLD activity in the presence of GTP␥S was inhibited by protein kinase C inhibitors (chelerythrine and calphostin C) and ADP-ribosylation of RhoA by C3 exoenzyme, indicating that the MDCK nuclear PLD activity is synergistically regulated via RhoA and protein kinase C. However, we have not observed synergy in the nuclear ARF-dependent PLD activities of both rat liver cells and AH7974 cells by either ATP or PMA when assayed with an exogenous substrate (PE/PIP 2 /PC vesicles), although considerable PKC␣ was present in these nuclei. C3 exoenzyme is known to block the binding to effector enzymes by ADP-ribosylating Rho proteins (RhoA, RhoB, RhoC) (36). Recently, Clostridium difficile toxin B has been found to be available for analyzing the role of Rho family proteins in PLD activation (37,38). Toxin B is monoglucosyltransferase catalyzing the incorporation of glucose into threonine residue at position 37 of Rho family proteins (Rho, Rac, Cdc42) (39). The ARF-dependent nuclear PLD activities of the hepatic cells were partially inhibited by toxin B, but not by C3 exoenzyme, suggesting that other Rho family protein rather than RhoA (most likely Cdc42) may be involved in the nuclear ARF-dependent PLD activity of rat hepatic cells (data not shown). Thus the rat liver nuclear PLD may be different from the MDCK nuclear enzyme in lower activation by PKC and insensitivity to C3 exoenzyme.
Several studies have shown that ARF plays an essential role in membrane trafficking by controlling the reversible assembly of the coat protein complex on the surface of Golgi membranes (40) and that ARF and ARF-stimulated PLD activity is enriched in Golgi membranes (18,41). On the other hand, Boman et al. (42) have suggested a role for ARF in nuclear vesicle dynamics during mitosis. Our results indicated that the specific activity of the nuclear ARF-dependent PLD activity, but not oleate-dependent enzyme, was transiently elevated in Sphase of rat regenerating liver cells. Moreover, the ARF-dependent PLD activity was extremely high in the nuclei of rat hepatoma AH7974 cells compared with that of the resting liver cells. The nuclear ARF levels was high in the AH7974 cells, and the increase of nuclear ARF level was observed in S-phase. Thus, one would speculate that the nuclear ARF-dependent PLD may be activated during S-phase to generate nuclear signaling molecules required for DNA synthesis or to enter into G 2 phase.
There are some possible explanations for the transient elevation of the nuclear ARF-dependent PLD activity during Sphase of rat liver regeneration. First, the GTP␥S-dependent PLD activity in nuclei is stimulated by translocation of cytosolic ARF to nuclei. This possibility could be supported by our finding that level of the nuclear ARF was increased in the S-phase and were higher in the AH7974 nuclei. Second, certain PLD is newly expressed by receptor activation by various stimulants produced after hepatectomy. The level of PLC␦4 in nuclei has been known to increase in liver regeneration (6). Finally, translocation of PLD to nuclei occurs in response to external stimuli induced by hepatectomy. Recent reports have demonstrated that cPLA 2 and MAP kinases also are translocated to nuclear envelope by cell stimulation with growth factors and A23187 (43)(44)(45). However, the mechanism underlying the transient elevation of the nuclear ARF-dependent PLD activity in the rat-regenerating liver remains to be disclosed.
A number of studies have provided evidence suggesting roles for the activation of nuclear PKC in nuclear functions involving phosphorylation of transcription factors and DNA replication factors (46). Also it has been shown that the activation of nuclear PKC is correlated with an increase in nuclear DG. Divecha et al. (8,10) have demonstrated that a turnover of nuclear PI is important in modulation of nuclear PKC in NIH3T3 cells. Furthermore, these notions are supported by the presence of nuclear PLC (3)(4)(5)(6). On the other hand, Jarpe et al. (12) have shown that the source of the thrombin-induced nuclear DG in fibroblasts is PC, not PI by examining the molecular species profiles. These reports suggest that both PI and PC contribute to the formation of nuclear DG, but the hydrolyzing enzymes are differently located in the nucleus; PLC is associated with the internal matrix (7), while the PC-hydrolyzing enzyme exists in the nuclear envelope (11). The result was supported by our results; the envelope-depleted nuclei from rat liver cells and AH7974 cells after treatment with Triton X-100 as described by Payrastre (7) had lesser activity of the nuclear ARF-dependent PLD, suggesting that the ARF-dependent PLD activity was located in the nuclear envelope (data not shown).
Many laboratories have demonstrated the nuclear localization of PKC isozymes in both resting and stimulated cells (11,(47)(48)(49). PKC␤ is located in the nuclei from unstimulated liver (47). We have demonstrated in this study that various levels of PKC isozymes, high levels of PKC␣ and PKC␤II and low level of PKC␦, were present in the isolated nuclei from rat resting liver cells and that the nuclear PKC␦ level, but not PKC␤II, was increased during S-phase. These results lead us to assume that nuclear translocation of the PKC␦ during S-phase may result from increase of nuclear DG levels produced by activation of the nuclear ARF-dependent PLD.
Acknowledgments-We thank Dr. Joel Moss (National Institutes of Health) for kind gifts of the ARF1 plasmid and the anti-ARF antibody FIG. 7. Western blot (A) and amount of G proteins (B) in the nuclei from AH7974, rat resting liver, and regenerating liver. The nuclei (300 g of protein) isolated from AH7974 (lane 1), rat resting liver (lane 2), and rat regenerating liver (28 h after partial hepatectomy) (lane 3) were separated on a 13% SDS-polyacrylamide gel electrophoresis and immunostained with the indicated antibodies. A blot representative of three independent experiments is shown. B, amount of G proteins of various nuclei shown in A was quantified by scanning densitometry and mass of all nuclei (AH cells, rat resting liver cells (Go), and rat regenerating liver cells (S)) of each G protein was designated as 100%. Results are shown as mean Ϯ S.E. from three independent experiments.