M phase of the cell cycle in HL-60 cells

Biochimica et Biophysica Acta 1631 (2003) 61 – 71 www.bba-direct.com

Nuclear phosphoinositide 3-kinase C2h activation during G2/M phase of the cell cycle in HL-60 cells Dora Visˇnjic´ a, Josip C´uric´ b, Vladiana Crljen a, Drago Batinic´ b, Stefano Volinia c, Hrvoje Banfic´ a,* a

Department of Physiology and Croatian Institute for Brain Research, School of Medicine, University of Zagreb, Salata 3, 10 000 Zagreb, Croatia b Department of Clinical Laboratory Diagnosis, Clinical Hospital Center, 10 000 Zagreb, Croatia c Dip. di Morfologia ed Embriologia, Universita degli Studi, Via Fossato di Mortara 64/b, 44100 Ferrara, Italy Received 4 July 2002; received in revised form 1 November 2002; accepted 14 November 2002

Abstract The activity of nuclear phosphoinositide 3-kinase C2h (PI3K-C2h) was investigated in HL-60 cells blocked by aphidicolin at G1/S boundary and allowed to progress synchronously through the cell cycle. The activity of immunoprecipitated PI3K-C2h in the nuclei and nuclear envelopes showed peak activity at 8 h after release from the G1/S block, which correlates with G2/M phase of the cell cycle. In the nuclei and nuclear envelopes isolated from HL-60 cells at 8 h after release from G1/S block, a significant increase in the level of incorporation of radiolabeled phosphate into phosphatidylinositol 3-phosphate (PtdIns(3)P) was observed with no change in the level of radiolabeled PtdIns(4)P, PtdIns(4,5)P2 and PtdIns(3,4,5)P3. On Western blots, PI3K-C2h revealed a single immunoreactive band of 180 kDa, whereas in the nuclei and nuclear envelopes isolated at 8 h after release, the gel shift of 18 kDa was observed. When nuclear envelopes were treated for 20 min with A-calpain in vitro, the similar gel shift and increase in PI3K-C2h activity was observed which was completely inhibited by pretreatment with calpain inhibitor calpeptin. The presence of PI3K inhibitor LY 294002 completely abolished the calpain-mediated increase in the activity of PI3K-C2h but did not prevent the gel shift. When HL-60 cells were released from G1/S block in the presence of either calpeptin or LY 294002, the activation of nuclear PI3K-C2h was completely inhibited. These results demonstrate the calpain-mediated activation of the nuclear PI3K-C2h during G2/M phase of the cell cycle in HL-60 cells. D 2002 Elsevier Science B.V. All rights reserved. Keywords: Phosphoinositide 3-kinase C2h; Nuclei; Calpain; Cell cycle; HL-60 cells

1. Introduction Several studies have indicated the role of nuclear phosphoinositide signalling distinct from the one in plasma membrane in the progression of cell cycle of mammalian cells. Early studies on mitogen-stimulated Swiss 3T3 cells or regenerating rat liver point to the role of specific increases in nuclear 1,2-diacylglycerol (DAG) level in cell proliferation [1,2]. When the nuclear phosphoinositide metabolism was followed in synchronized cells, which allow more precise definition of particular stages of cell cycle, two different points were described to be associated with a specific increase in the level of nuclear DAG. In * Corresponding author. Zavod za Fiziologiju, Medicinski Fakultet, Sveucˇilisˇte u Zagrebu, Sˇalata 3, POB 978, 10 001 Zagreb, Croatia. Tel.: +385-1-4590-260; fax: +385-1-4590-207. E-mail address: [email protected] (H. Banfic´).

synchronized HeLa and MEL cells, a decrease in the level of nuclear phosphatidylinositols (PtdIns) or an increase in PtdIns-phospholipase C (PI-PLC) activity was observed either during the S phase or somewhere at the G1/S transition [3,4]. In HL-60 cells synchronized by aphidicolin, an increase in the level of DAG and the activity of PIPLC was detected in the nuclei isolated at G2 phase, and the activity of PI-PLC and progression through the G2/M boundary was inhibited by a specific PI-PLC inhibitor [5]. Moreover, the similar cell cycle arrest in G2 phase was observed when the cells were treated with chelerethrine chloride, an inhibitor of hII protein kinase C (PKC) isoform which was shown to phosphorylate lamin B in HL-60 cells [6,7]. These studies suggested the role of PIPLC-mediated increase in the nuclear DAG in the activation of PKC hII which is by phosphorylating lamin B responsible for nuclear lamina disassembly and entry into mitosis.

1388-1981/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. doi:10.1016/S1388-1981(02)00356-6

62

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

Phosphoinositide 3-kinases (PI3K) of class IA and class IB have been recently reported to translocate into the nuclei of cells stimulated with serum or growth factors, although their role in the progression of cell cycle was not investigated. PI3-kinases are a family of enzymes that share a homologous region in the catalytic subunit and are divided into three distinct classes on the basis of their in vitro substrate specificity, structure and mode of regulation [8]. The most studied are class I PI3Ks, which generate 3phoshorylated phosphoinositides in response to different extracellular stimuli; although these enzymes in vitro phosphorylate PtdIns, PtdIns(4)P, and PtdIns(4,5)P2, they display a strong preference for PtdIns(4,5)P2 in vivo. The class IA enzymes consists of a 110-kDa catalytic subunit and an 85kDa adaptor protein which links the enzyme to tyrosine kinases in cell membranes; the presence of both p110h catalytic and p85a regulatory subunit was detected in cell nuclei [9,10]. The class IB enzyme is composed of p110g catalytic subunit and a p101 subunit regulated by G proteins; the presence of G protein-activated PI3Kg was demonstrated in the nuclei of cells induced to proliferate [11,12]. The mammalian class III enzyme is highly related to yeast Vps34 gene product and, like the yeast enzyme, is specific for PtdIns and will not phosphorylate PtdIns(4)P or PtdIns (4,5)P2 [13]. Little is known about the mode of activation of class II PI3K enzymes which include three isoforms; PI3K-C2a and -h are fairly ubiquitous, in contrast to PI3K-C2g which is restricted to liver. These enzymes are of high molecular mass (>170 kDa), their defining feature is the presence of a Phox homology (PX) and C2 domain at their C-termini and they where shown to phosphorylate both PtdIns and PtdIns(4)P in vitro [14 – 17]. Our recent studies showed that the peak of rat liver regeneration after partial hepatectomy was associated with an increase in the level of nuclear PtdIns(3)P and the calpain-mediated activation of PI3K-C2h in the membrane-depleted nuclei [18], while in all-trans retinoic acid (ATRA)-differentiated HL-60 cells, nuclear PI3K-C2h is activated by tyrosine phosphorylation [19]. In the present study, we investigated the metabolism of 3phosphorylated phosphoinositides in a well-described model of aphidicolin-treated HL-60 cells which allows the cells to progress synchronously through the cell cycle. Evidence is provided that the activation of PI3K-C2h and the increase of PtdIns(3)P occurs at G2/M phase of the cell cycle. Furthermore, we show that PI3K-C2h activation in the nuclei and nuclear envelopes may be a calpain-mediated event.

2. Materials and methods 2.1. Materials Reagents were obtained from the following sources: EGTA, EDTA, HEPES, Tris, leupeptin, phenylmethylsul-

fonyl fluoride, phosphatidylserine (PtdSer), PtdIns, DNase I, RNase A, aphidicolin, Triton X-100, Na+ deoxycholate, protein A-Sepharose, SDS, fetal bovine serum (FBS) and aprotinin from Sigma, St. Louis, MO, USA; LY 294002, Acalpain and calpeptin from Calbiochem, Nottingham, UK; [g-32P]ATP, and enhanced chemiluminescence kit from Amersham Pharmacia Biotech, Amersham, Bucks, UK; Antibodies to A-calpain (preautolytic and postautolytic) were a generous gift from Dr. C. Saido, RIKEN Brain Science Institute, Saitama, Japan; anti-PI3K p85 antibody from Upstate Biotechnology, Lake Placid, NY, USA; antiPI3K p110g and anti-lamin B antibodies from Santa Cruz Biotechnology, Santa Cruz, CA, USA. All other chemicals were of analytical grade. 2.2. Cell culture, cell cycle synchronization, and flow cytometric analysis HL-60 cells (ECCACC no. 88112501) were obtained from the European Collection of Animal Cell Cultures, PHLS, Porton, Salisbury, UK. The cells were maintained in exponential growth in RPMI 1640 medium with 10% heat-inactivated FBS, 100 U/ml penicillin and 100 Ag/ml streptomycin in a 5% CO2 humidified atmosphere at 37 jC. For cell cycle synchronization, cells were treated with 2 Ag/ ml aphidicolin at a cell density of 0.8  106/ml for 18 h. Cells were removed from G1/S blockade (time 0) by washing four times with RPMI and resuspending the cells in RPMI containing 10% FBS. In some cases, the PI3K inhibitor LY 294002 or calpain inhibitor calpeptin were added to the cultures at the times and concentrations indicated in the figure and table legends. Cells were sampled at the indicated time points after release from aphidicolin, fixed in 75% ethanol and stored at 20 jC for at least 24 h. Following this, the cells were centrifuged at 800  g for 15 min and supernatants were discarded to remove ethanol completely. The pellets were resuspended in 40 Al (for 2 – 3  106 cells) of PBS at room temperature for 30 min, washed and stained with propidium iodide solution (20 Ag/ml propidium iodide and 20 Ag/ml RNase A in PBS) for 30 min. Cell cycle analysis was performed using a FACScan flow cytometer as described previously [5]. 2.3. Isolation of nuclei, membrane-depleted nuclei, postnuclear membranes, cytosolic fraction and nuclear envelopes and determination of their purity At the indicated time points, cells were sampled, washed three times with cold PBS and resuspended in 50 mM Tris – HCl, pH 7.4, 250 mM sucrose, 5 mM MgSO4 (STM buffer) containing 0.1 mM sodium metavanadate (NaVO3), 20 mM sodium fluoride, 10 Ag/ml leupeptin, 1 mM phenylmethylsulfonyl fluoride and 1% (v/v) 2-mercaptoethanol. After 5 min, the cells were lysed with 20 strokes of a power-driven teflon pestle. The lysate was layered over a cushion of 2.1

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

M sucrose, 50 mM Tris – HCl, pH 7.4, 5 mM MgSO4 containing inhibitors in the same concentration as STM buffer. The samples were then spun at 70,000  g for 60 min in a SW 28.1 rotor. The pellet at the bottom of the cushion was considered to be the nuclear fraction [20]. Preparation of membrane-depleted nuclei was performed with the non-ionic detergent Triton X-100. A 0.5 ml aliquot of nuclei in STM buffer was mixed with 20 ml of ice-cold buffer containing 5 mM Tris (pH 7.4), 5 mM MgCl2, 1.5 mM KCl, 1 mM EGTA, 0.29 M sucrose and left for 20 min. Triton X-100 was added to this buffer before addition of the nuclei to yield final concentration of 0.04% (w/v). The nuclei were pelleted at 165  g for 6 min (4 jC) and the supernatant removed. The pellet was carefully resuspended in STM buffer (5 ml), and a cushion carefully laid beneath it [10 mM HEPES (pH 7.5), 2 mM MgCl2, 0.5 M sucrose; 10 ml], and centrifuged at 165  g for 6 min (4 jC). This was assumed to remove any residual Triton X-100. The supernatant was removed and the final pellet resuspended in 0.5 ml of STM buffer [18]. Preparation of postnuclear membranes was achieved by centrifugation of the supernatant, which remained above cushion after the nuclear fraction was obtained. The supernatant was diluted in STM buffer to give a final concentration of 162 mM sucrose and spun at 106,000  g for 90 min at 4 jC in a Beckman SW 28.1 rotor. The resultant pellet was considered to contain postnuclear membranes [18]. Cytosolic fraction was prepared by homogenization of cells as described above and afterwards samples were spun at 106,000  g for 90 min at 4 jC in a Beckman SW 28.1 rotor and the clear supernatant was considered to be cytosolic fraction. Nuclear envelopes were isolated from purified nuclei as described previously [20]. Briefly, purified nuclei were resuspended in STM buffer containing 1% (v/v) 2-mercaptoethanol at 1  108 nuclei/ml and incubated for 1 h at 4 jC with 100 Ag/ml DNase I and 100 Ag/ml RNase A. After the incubation, nuclei were sedimented at 800  g for 10 min and resuspended in 50 mM Tris – HCl, pH 7.4, at 5  108 nuclei/ ml; and 4 volumes of 50 mM Tris – HCl, pH 7.4, containing 2 M NaCl and 1% (v/v) 2-mercaptoethanol were added dropwise. After a 30 min incubation, nuclear envelopes were recovered by sedimentation at 5000  g for 30 min. The activities of cytochrome oxidase and glucose-6phosphatase were determined according to Wakabayashi et al. [21] and Nordlie and Arion [22], respectively, while the activity of 5V-nucleotidase was measured as described by Burnside and Schneider [23]. 2.4. Labeling of inositol lipids with [c-32P]ATP For the in vitro labeling of inositol lipids with [g-32P]ATP nuclei or nuclear envelopes (total protein 100 Ag) were resuspended in 90 Al of buffer containing 10 mM HEPES (pH 7.5), 5 mM MgCl2, 1.5 mM KCl, 1 mM EGTA, and 0.25 M sucrose. The samples were preincubated for 2 min at 30 jC

63

to hydrolyze any remaining endogenous ATP. Then 10 Al of phosphorylation mixture (40 ACi of [g-32P]ATP, 2 Al of 5 mM non-radiolabeled ATP, made up to 10 Al with the abovementioned buffer) was added. Incubation was carried out for 5 min at 30 jC and terminated by the addition of 1 ml of chloroform/methanol (1:1). Lipids were extracted and deacylated, and the separation of all the glycerophosphoinositides was achieved using an HPLC high resolution 5 AM Partisphere SAX column (Whatman) with a discontinuous gradient up to 1 M (NH4)2HPO4  H2PO4 (pH 3.8) exactly as described in a previous study [18]. 2.5. Immunoprecipitation of PI3K-C2b PI3K-C2h isoform-discriminant polyclonal antisera against the first 331-amino acid portion of PI3K-C2h [15], expressed in Escherichia coli as an amino-terminally fused glutathione S-transferase protein, were raised in rabbits as described previously [16,18,24]. These antisera were used for all immunoprecipitations and Western blots directed at PI3KC2h. Purified nuclei, postnuclear membranes, and nuclear envelopes were resuspended in 0.5 ml of buffer containing 50 mM Tris (pH 7.6), 150 mM NaCl, 1% Triton X-100 (w/v), 0.5% Na+ deoxycholate (w/v), 0.1% SDS (w/v), 2 mM phenylmethylsulfonyl fluoride, 1 Ag/ml aprotinin, and 1 Ag/ ml leupeptin and spun at 100,000  g for 90 min at 4 jC. PI3K-C2h was immunoprecipitated overnight from 450 Al of supernatants with antibody and protein A-Sepharose. Immunoprecipitates were washed once with the abovementioned buffer, then three times with 5 mM HEPES/2 mM EDTA (pH 7.5) and then the phosphorylation assay was carried out as described above, except that immunoprecipitates were resuspended in 40 Al of buffer containing 10 mM HEPES (pH 7.5), 5 mM MgCl2, 1.5 mM KCl, 1 mM EGTA, and 0.25 M sucrose and made up to 90 Al with lipid vesicles which consisted of 50 mM PtdIns and 100 mM PtdSer formed by sonication. Incubation was terminated, lipids were extracted, deacylated and separation of glycerophosphoinositide was achieved as described above [18]. 2.6. Western blot analysis Proteins for electrophoresis were prepared so that the concentration of each sample was 50 Ag/25 Al of sample loading buffer [25], and electrophoresis was carried out using a Bio-Rad minigel apparatus at an acrylamide concentration of 5% (w/v) or 10% (w/v) when the presence of lamin B, p85a and A-calpain was determined. After electrophoresis, the proteins were transferred to nitrocellulose using a Bio-Rad wet-blotting system. The blot was blocked with a buffer containing 4% (w/v) dried milk, 20 mM Tris, 140 mM NaCl, 0.05% (v/v) Tween 20. It was then probed for 2 h with primary antibody (1:1000), then washed with a blocking buffer and incubated with the secondary antibody conjugated to horseradish peroxidase. Visualization was carried out using the ECL kit.

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

64

2.7. Statistical evaluation The data are shown as means F S.E. For statistical analyses, the Student’s t-test for unpaired samples at the level of significance of 0.05 was used. Fig. 1. Western blot analysis of lamin B in subcellular fractions. Protein (50 Ag) was subjected to SDS-PAGE, transferred to nitrocellulose, and probed with anti-lamin B antibody: lane 1, nuclei; lane 2, postnuclear membranes. The position of molecular mass marker for albumin, bovine serum (66 kDa) is indicated on the left side by the arrow.

3. Results 3.1. Assessment of purity of subcellular fractions The purity of subcellular fractions was assessed by measurement of activity of marker enzymes for microsomes (glucose-6-phosphatase), plasma membrane (5V-nucleotidase) and mitochondria (cytochrome oxidase). As shown in Table 1, isolated nuclei were free of plasma membrane and mitochondrial marker enzymes with only residual contamination with microsomal marker enzyme (glucose-6phosphatase), which was also observed in our previous studies [2,18] when liver nuclei were purified even in the presence of strong detergents, suggesting that microsomal marker enzymes are present in the nuclear membrane. When compared to homogenate, all marker enzymes were enriched in postnuclear membranes, while in cytosolic fraction, no activity of marker enzymes could be found. Moreover, when the presence of lamin B was investigated in the nuclei and postnuclear membranes (Fig. 1), lamin B could only be found in the nuclei. Altogether, the above shown data clearly demonstrate that subcellular fractions were purified to a satisfactory degree and could be used for further biochemical studies. 3.2. The activity of PI3K-C2b in nuclei and nuclear envelopes of HL-60 cells changes during the cell cycle To allow synchronous progression through the cell cycle, HL-60 cells were arrested at the G1/S boundary by treatment with aphidicolin for 18 h, washed and released into the

Table 1 Determination of marker enzymes in subcellular fractions Fraction

Glucose-6phosphatasea

5V-nucleotidasea

Cytochrome oxidaseb

Homogenate Nuclei Postnuclear membranes Cytosolic fraction

0.324 F 0.041 0.035 F 0.007 0.872 F 0.087

0.117 F 0.011 n.d. 0.197 F 0.025

62.1 F 3.9 n.d. 155.2 F 9.8

n.d.

n.d.

n.d.

Marker enzymes for microsomes (glucose-6-phosphatase), plasma membrane (5V-nucleotidase) and mitochondria (cytochrome oxidase) were measured in subcellular fractions as described in Materials and methods. The results are means F S.E. for three different experiments, each performed in duplicate. n.d., not detectable. a Amol Pi/mg protein per hour. b n atoms oxygen/mg protein per min.

medium. An aliquot of cells was harvested at indicated times, stained with propidium iodide and flow cytometric analysis showed that the cells were synchronously progressing through S phase, G2 phase, mitosis and the subsequent G1 phase (Fig. 2A). Nuclei were isolated through the sucrose cushion from cells at different times after the release from G1/S block and assayed for the activity of immunoprecipitated PI3K-C2h. The nuclear level of PI3K-C2h in G1 phase cells is comparable with that in unsynchronized cells, indicating that the synchronization procedure itself did not affect nuclear PI3K-C2h levels (data not shown). No changes in the activity of immunoprecipitated PI3K-C2h were observed in the nuclei isolated from cells at 30 min to 4 h after G1/S block. A significant increase in the level of PI3K-C2h activity was observed at 6 h after the release, the peak activity of the enzyme was measured at 8 h, the activity was still significantly increased at 12 h and then returned to the baseline levels at 18 h (Fig. 2B). No increase in the level of immunoprecipitable PI3K-C2h activity was observed in postnuclear membranes at any time point tested (Fig. 2B), or cytosolic fraction (data not shown). Moreover, when the nuclei at the peak of the enzyme activity (8 h) were washed with Triton X-100 (0.04%), which is known to remove the nuclear envelope [18,26], the enzyme activity dropped from 4692 F 278 (dpm of PtdIns(3)P/100 Ag of protein, n = 4) to 1201 F 84 (dpm of PtdIns(3)P/100 Ag of protein, n = 4), suggesting that the majority of the enzyme is localized in the nuclear envelope. Therefore, nuclear envelopes were prepared using the sequential treatment of the nuclei with nucleases and 1.6 M NaCl buffer containing 2mercaptoethanol. The procedure was previously shown to quantitatively remove nucleic acids, histones and some nonhistone proteins, resulting in the preparation of highly purified nuclear envelopes from HL-60 cells [20]. As shown in Table 2, the activity of PI3K-C2h in nuclear envelopes isolated from HL-60 cells at 8 h after the release from G1/S block was significantly increased in comparison to the activity of the enzyme in nuclear envelopes immediately after the release. The increase in the activity of immunoprecipitated PI3K-C2h from nuclear envelopes of HL-60 cells at 8 h after the release from G1/S block was completely inhibited when the phosphorylation assay was performed in the presence of PI3K inhibitor LY 294002 (50 AM). It is important to note, that the abovementioned sensitivity of the

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

65

Fig. 2. (A) Flow cytometric analysis of the cell cycle progression of HL-60 cells synchronized by block and the release from G1/S. HL-60 cells were synchronized in the G1/S phase as described in Materials and methods and allowed to progress synchronously through the cell cycle. Cells were harvested at the indicated times after the release from G1/S phase, stained with propidium iodide and assessed for cell cycle distribution by flow cytometric analysis. (B) Activity of immunoprecipitated PI3K-C2h in nuclei and postnuclear membranes of synchronized HL-60 cells at different time points after the release from G1/S. Nuclei ( ) and postnuclear membranes (o) were prepared as described in Materials and methods. Kinase assay was performed using PtdIns as substrate, and all other details are as described in Materials and methods. Results are means F S.E. for three different experiments, each performed in duplicate. *P < 0.05 (Student’s t-test) with respect to the control.

.

immunoprecipitable PI3K-C2h activity to the LY 294002 (50 AM) was similar to the value observed when dosedependent analysis of the sensitivity of the immunoprecipitable PI3K-C2h activity towards LY 294002 was investigated [19] and is comparable to that which was observed when inhibitor was tested on the purified recombinant PI3K-C2h activity [16].

3.3. Phosphorylation of phosphoinositides and proteolytic activation of PI3K-C2b in the nuclei and nuclear envelopes isolated from HL-60 cells at 8 h after the release from G1/S block Nuclei (Fig. 3A) and nuclear envelopes (Fig. 3B) were prepared from HL-60 cells harvested at 8 h or immediately

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

66

Table 2 Activity of immunoprecipitated PI3K-C2h in the nuclear envelopes of synchronized HL-60 cells

PI3K-C2h activity (dpm of PtdIns(3)P/ 100 Ag of protein)

0h

8h

8 h LY 294002

2232 F 88

6383 F 274*

2170 F 98

Nuclear envelopes were isolated from synchronized HL-60 either immediately or 8 h after the release from the G1/S block. Kinase assay was performed using PtdIns as substrate and all other details are as described in Materials and methods. When added PI3K inhibitor LY 294002 (50 AM) was present only during phosphorylation assay. The results are means F S.E. for three different experiments, each performed in duplicate. * P < 0.05 (Student’s t test) with respect to the control.

after the G1/S block and labeled with [g32P]ATP for 5 min. HPLC analysis of deacylated phospholipids extracted from nuclei and nuclear envelopes showed the incorporation

Fig. 3. Incorporation of 32P into PtdIns(3)P, PtdIns(4)P, PtdIns(4,5)P2, and PtdIns(3,4,5)P3 in the nuclei (A) and nuclear envelopes (B) isolated from synchronized HL-60 cells. Nuclei (A) and nuclear envelopes (B) were isolated from synchronized HL-60 cells either immediately (open bars) or 8 h after the release from the G1/S phase (black bars). The nuclei and nuclear envelopes were radiolabeled, and the separation of glycerophosphoinositides was achieved as described in Materials and methods. The effect of the presence of LY 294002 (50 AM) during the phosphorylation assay in the nuclei and nuclear envelopes harvested at 8 h after the release from the G1/S phase is shown (gray bars). The results are means F S.E. for three different experiments, each performed in duplicate. *P < 0.05 (Student’s t-test) with respect to the control.

of 32P in PtdIns(3)P, PtdIns(4)P, PtdIns(4,5)P2 and PtdIns (3,4,5)P3. In the nuclei and nuclear envelopes isolated from HL-60 cells at 8 h after the release from G1/S block, a significant increase in the level of radiolabeled PtdIns(3)P was observed which was completely inhibited by the presence of LY 294002 (50 AM). The level of incorporation of 32 P into PtdIns(4)P, PtdIns(4,5)P2 and PtdIns(3,4,5)P3 in the nuclei and nuclear envelopes at 8 h after the release from aphidicolin block did not differ from the level of incorporation in phospholipids in the control nuclei and nuclear envelopes. Western blotting of postnuclear membranes, nuclei and nuclear envelopes isolated at time 0, fractionated with SDSPAGE on a 5% gel and probed with antisera raised against PI3K-C2h, revealed a single immunoreactive band of 180 kDa (Fig. 4). In the nuclei and nuclear envelopes isolated from cells at 8 h after the release from G1/S block, a gel shift of 18 kDa was observed, while no change in either the size or the amount of immunoreactive protein was detected in postnuclear membranes. On the other hand, no difference in the amount of p85a immunoreactive protein in the nuclei and nuclear envelopes isolated from the control cells and cells at 8 h after G1/S block could be observed (results not shown). Moreover, when phosphorylation assay was carried out in the presence of Ca2 + instead of Mg2 +, which would improve the specificity of the assays towards that of the class II PI3K enzymes as described by Arcaro et al. [16], no difference in the incorporation rate of 32P into PtdIns(3)P could be observed (Table 3) further ruling out a possible involvement of class I PI3K in the increase of nuclear PtdIns(3)P. It is important to note that no immunoreactive p110g was present in either nuclei or nuclear envelopes (data not shown). The calpains are predominantly cytoplasmic Ca2 +dependent proteases, but it has been observed that A-calpain can be transported into the cell nuclei in an ATP-dependent fashion [27], where it causes proteolysis of nuclear proteins [28]. As shown in Fig. 5. A-calpain is present in the nuclei of HL-60 cells and at 8 h after release of the cells from G1/S block it undergoes partial proteolysis which is known to be autolytic [29] and which coincide with the activation and

Fig. 4. Western blot analysis of PI3K-C2h in subcellular fractions of synchronized HL-60 cells. Postnuclear membranes (PNM), nuclei (N) and nuclear envelopes (NE) were isolated from synchronized HL-60 either immediately (0 h) or 8 h after the release from the G1/S phase (8 h). Protein (50 Ag) was subjected to SDS-PAGE, transferred to nitrocellulose, and probed with anti-PI3K-C2h antibody. The position of the molecular mass marker for a2-macroglobulin (180 kDa) is indicated on the left side by the arrow.

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

67

Table 3 Incorporation of 32P into PtdIns(3)P in the nuclei and nuclear envelopes isolated from synchronized HL-60 cells Incorporation of of protein

Control nuclei Nuclei (8 h) Control nuclear envelopes Nuclear envelopes (8 h)

32

P into PtdIns(3)P dpm/100 Ag

Mg2 +

Ca2 +

311 F 56 1278 F 173* 372 F 69

298 F 61 1219 F 151* 369 F 75

1496 F 152*

1538 F 129*

Nuclei and nuclear envelopes were isolated from synchronized HL-60 cells either immediately or 8 h after the release from the G1/S phase. The nuclei and nuclear envelopes were radiolabeled, and the separation of glycerophosphoinositides was achieved as described in Materials and methods. When phosphorylation assay was carried out in the presence of Ca2 +, 5 mM CaCl2 was added to the buffer instead of 5 mM MgCl2 and 1 mM EGTA was replaced by 1 mM EDTA. The results are means F S.E. for three different experiments, each performed in duplicate. * P < 0.05 (Student’s t-test) with respect to the control.

proteolytic cleavage of PI3K-C2h in the cell nuclei and nuclear envelopes (Figs. 2 and 4 and Table 2). We have previously shown that a similar pattern of activation of PI3K-C2h in liver nuclei and platelets could be prevented by calpain inhibitor calpeptin [18,24]. To further investigate the role of calpain in PI3K-C2h activation in HL-60 cells, the nuclear envelopes were isolated from HL-60 cells synchronized in G1/S phase. The short-term exposure of nuclear envelopes to calpain in vitro significantly increased the level of immunoprecipitated PI3K-C2h activity (Fig. 6A) and caused the gel shift of immunoreactive protein (Fig. 6B) similar to the one observed in nuclear envelopes isolated from cells at 8 h after the release from aphidicolin block (Fig. 4). The incubation of nuclear envelopes with 50 AM Ca2 + alone had no effect on either the activity of PI3KC2h or the immunoreactive protein. The pretreatment of nuclear envelopes with calpain inhibitor calpeptin completely abolished the effects of calpain on both the activity of enzyme and the gel shift. Although the presence of PI3K inhibitor LY 294002 completely abolished the calpain-

Fig. 6. Effect of calpain on immunoprecipitable PI3K-C2h activity (A) and their Western blot analysis (B) in the nuclear envelopes of HL-60 cells in vitro. Nuclear envelopes were isolated from HL-60 cells and treated with Acalpain (15 Ag/ml) at 25 jC for 20 min in the presence of 50 AM free Ca2 + concentration. Calpain inhibitor calpeptin (200 Ag/ml) or PI3K inhibitor LY 294002 (50 AM) were added prior to the exposure of nuclear envelopes to calpain. After the incubation, nuclear envelopes were washed three times in STM buffer, and immunoprecipitable PI3K-C2h activity was measured using PtdIns as a substrate (A). For Western blot analysis (B), the same experiments were performed, and protein (50 Ag) was subjected to SDSPAGE, transferred to nitrocellulose, and probed with anti-PI3K-C2h antibody: lane 1, control; lane 2, calpain-treated nuclear envelopes; lane 3, nuclear envelopes incubated with 50 AM Ca2 + alone; lane 4, nuclear envelopes treated with calpain in the presence of calpeptin; lane 5, nuclear envelopes treated with calpain in the presence of LY 294002. The position of the molecular mass marker for a2-macroglobulin (180 kDa) is indicated on the left side by the arrow.

mediated increase in the activity of immunoprecipitated PI3K-C2h in phosphorylation assay, the PI3K inhibitor had no effects on calpain-induced gel shift of immunoreactive protein (Fig. 6A and B). On the other hand, when intact nuclei were treated with calpain as described above, no increase in the level of immunoprecipitated PI3K-C2h activity (Fig. 7A) and no gel shift of immunoreactive protein (Fig. 7B) could be observed, suggesting that calpaininduced activation of the enzyme takes place at the inner surface of the nuclear envelope. 3.4. Treatment of HL-60 cells with calpeptin and LY 294002 prevent the activation of nuclear PI3K-C2b

Fig. 5. Western blot analysis of A-calpain in the nuclei of synchronized HL60 cells. Nuclei were isolated from synchronized HL-60 cells either immediately or 8 h after the release from the G1/S phase. Protein (50 Ag) was subjected to SDS-PAGE, transferred to nitrocellulose, and probed with anti- (preautolytic and postautolytic) A-calpain antibodies: lane 1, control nuclei; lane 2, nuclei harvested 8 h after the release from the G1/S phase. The position of molecular mass marker for fructose-6-phosphate kinase (84 kDa) is indicated on the left side by the arrow.

To test for the possible role of calpeptin and LY 294002 on the nuclear PI3K-C2h activation, HL-60 cells were synchronized for 18 h with aphidicolin and then washed and resuspended in either the control medium or the medium containing calpeptin (200 Ag/ml) or LY 294002 (50 AM). Nuclear envelopes were harvested 8 h after release

68

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

from G1/S and assayed for PI3K-C2h activity and the gel shift. As shown in Fig. 8A, the increase in the activity of PI3K-C2h at 8 h after the release from the block was completely inhibited when the cells were allowed to progress in the presence of either calpeptin or LY 294002. The presence of calpeptin during the last 60 min of the synchronization procedure, and for 8 h after the release, completely inhibited both the activity of PI3K-C2h in phosphorylation assay and the gel shift that was observed in nuclei of the control cells (Fig. 8B). In contrast, the treatment of cells with LY 294002 for 8 h after release from G1/S block completely inhibited the activity of PI3K-C2h but the gel shift on Western blot analysis was similar to the one observed in the control cells (Fig. 8A and B). In addition, nuclear envelopes were isolated from the cells released into the control medium for 8 h and then treated for 5 min with either LY 294002 or calpeptin in vitro. The presence of LY 294002 in a phosphorylation assay completely inhibited the increase in PI3K-C2h activity while having no effects on the gel shift. In contrast, the presence of calpeptin in vitro had no effects on either the PI3K-C2h activity or the gel shift further suggesting that the increase in the PI3K-C2h

Fig. 8. The effects of calpeptin and LY 294002 on the activity of immunoprecipitable PI3K-C2h (A) and their Western blot analysis (B) in the nuclear envelopes of synchronized HL-60 cells in vivo and in vitro. HL60 cells were synchronized in the G1/S phase and released into the control medium (open bars) or the medium containing either 50 AM LY 294002 or 200 Ag/ml calpeptin (gray bars). Nuclear envelopes were harvested 8 h after the release from G1/S and assayed for kinase activity using PtdIns as substrate as described in Materials and methods. For an aliquot of nuclear envelopes isolated from cells 8 h after the release into the control medium, the phosphorylation assay was performed in the presence of either 50 AM LY 294002 or 200 Ag/ml calpeptin in vitro (black bars). The results are means F S.E. for three different experiments, each performed in duplicate. *P < 0.05 (Student’s t-test) with respect to the control. For Western blot analysis (B), the same experiments were performed, and protein (50 Ag) was subjected to SDS-PAGE, transferred to nitrocellulose, and probed with anti-PI3K-C2h antibody: lane 1, nuclear envelopes isolated from cells at time 0; lane 2, nuclear envelopes isolated at 8 h after the release of cells into the control medium; lane 3, nuclear envelopes isolated at 8 h after the release of cells into the medium containing 50 AM LY 294002; lane 4, nuclear envelopes isolated at 8 h after the release of cells into the medium containing 200 Ag/ml calpeptin; lane 5, nuclear envelopes isolated at 8 h after the release of cells into the control medium and then treated for 5 min with 50 AM LY 294002 in vitro; lane 6, nuclear envelopes isolated at 8 h after the release of cells into the control medium and then treated for 5 min with 200 Ag/ml calpeptin in vitro. The position of the molecular mass marker for a2-macroglobulin (180 kDa) is indicated on the left side by the arrow.

activity in the nuclear envelopes isolated at 8 h after the release into the control medium was a calpain-mediated event. Fig. 7. Effect of calpain on immunoprecipitable PI3K-C2h activity (A) and their Western blot analysis (B) in the intact nuclei of HL-60 cells in vitro. Intact nuclei were isolated from HL-60 cells and treated with A-calpain (15 Ag/ml) at 25 jC for 20 min in the presence of 50 AM free Ca2 + concentration. After the incubation, nuclei were washed three times in STM buffer, and immunoprecipitable PI3K-C2h activity was measured using PtdIns as a substrate (A). For Western blot analysis (B), the same experiments were performed, and protein (50 Ag) was subjected to SDSPAGE, transferred to nitrocellulose, and probed with anti-PI3K-C2h antibody: lane 1, control nuclei; lane 2, calpain-treated nuclei. The position of the molecular mass marker for a2-macroglobulin (180 kDa) is indicated on the left side by the arrow.

4. Discussion Although it has been recognized that class II phosphoinositide 3-kinases are downstream targets for a growing number of receptors located in the plasma membrane, little is known about the physiological role and mechanism of activation of either PI3K C2a or PI3K-C2h. PI3K-C2a plays a signalling role downstream of both the receptor for

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

the chemokine monocyte chemotactic peptide (MCP-1) and the insulin receptor [30,31], PI3K-C2h is activated in platelets in response to stimulation of integrin receptors by fibrinogen [24] and both enzymes are downstream signalling targets of activated epidermal growth factor (EGF) and platelet-derived growth factor (PDGF) receptors [32]. In contrast to the class I enzymes which are mainly cytosolic and which are widely considered to be involved in the signal transduction at plasma membrane, class II PI3K activities are predominantly associated with the membrane fraction of cells. The activity of PI3K-C2h was found to be constitutively associated with the plasma membrane and low-density microsomal fraction [16], different immunofluorescence studies showed that PI3K-C2a is concentrated in either trans-Golgi network and clathrin-coated vesicles [33], or clathrin-coated vesicles only and nuclear speckles [34], and the expression of epitope-tagged rat PI3K-C2g suggested that this class II PI3K isozyme is present on the plasma membrane and nuclear membranes [17]. Our previous work has shown that the activity of immunoprecipitable PI3KC2h was present in membrane-depleted rat liver nuclei and significantly increased during the compensatory liver growth [18]. Using the model of HL-60 cells synchronized by aphidicolin, the results of the present study suggest that the progression of cells through G2/M boundary was associated with a significant increase in the activity of immunoprecipitable PI3K-C2h activity in the nuclei and nuclear envelopes with no changes in the level of the enzyme activity in either cytosolic fraction or postnuclear membranes. Polyclonal antisera against the first 331-amino acid portion of PI3K-C2h used in immunoprecipitation and Western blot analysis were previously shown to be specific for PI3K-C2h; antisera do not bind to either class I PI3Ks, PI3K-C2a, PI3K-C2g or Vps34p [16,18,24]. A strong preference for PtdIns as a substrate over PtdIns(4)P or PtdIns(4,5)P2 and sensitivity to LY 294002 further distinguished the activity of PI3K-C2h from the activity of class I or PI3K-C2a kinases, respectively [8]. The present study further corroborates the finding that the activation of PI3K-C2h might be, at least in some instances, a calpain-mediated effect: (1) the increase in the activity of PI3K-C2h in the nuclei and nuclear envelopes was associated with an 18 kDa gel shift of immunoreactive protein suggesting a proteolytic activation of the enzyme, (2) a similar increase in the activity and gel shift was observed when the nuclear envelopes were treated with calpain in vitro, and the effect was abolished by pretreatment of envelopes with calpeptin, and (3) the treatment of cells with calpeptin inhibited the increase in the activity of PI3K-C2h in the nuclear envelopes isolated from the cells at 8 h after the release from G1/S block. The same gel shift of PI3K-C2h and the sensitivity to calpain inhibitors was previously observed in the platelets, where PI3K-C2h was activated in response to integrin receptor stimulation by fibrinogen [24]. In the membrane-depleted rat nuclei, a short-time exposure of enzyme to calpain was shown to

69

result in similar degree of activation compared to the nuclei isolated 20 h after partial hepatectomy [18]. A possibility of activation of the enzyme with proteolytic degradation was suggested first by Arcaro et al. [16] who showed that the lipid kinase activity of a newly cloned PI3K-C2h isoform increased after a deletion of C2 domain. As the binding of PtdIns to C2 domain was shown to inhibit enzyme activity, the model was proposed in which the C2 domain of PI3K-C2h functions as a negative regulator of the lipid kinase activity of the enzyme, by directly competing with the catalytic domain for binding to PtdIns substrate vesicles [16]. From the amino acid sequence of the enzyme, it could be deduced that calpain-mediated proteolysis may cleave the C2 domain, which might be the mechanism responsible for the 18 kDa gel shift and an associated increase in the activity of PI3K-C2h we observed in the present study, as well as in the fibrinogen-stimulated platelets [24] or regenerating liver nuclei [18]. The calpains are predominantly cytoplasmic Ca2 +dependent proteases, but it has been observed that Acalpain can be transported into the cell nuclei in an ATPdependent fashion [27]. Although the proteolytic activity of two calpain isoforms differ in their sensitivity to Ca2 + concentration, m-calpain being active at millimolar Ca2 + concentrations, and A-calpain at micromolar Ca2 + concentrations, both isoforms are able to process cellular substrates at physiological Ca2 + concentration [29]. Recently, some adaptors, different from those that associate with the class I or class III enzymes, were found to bind to Nterminal region and activate the class II PI3K enzymes. Proline-rich motifs in N-terminal region of PI3K-C2h bind Grb2 adaptor molecule that stimulates the catalytic activity and mediates the association of this enzyme with activated EGF receptor [35]. Clathrin functions as an adaptor for the class II PI3K-C2a binding to its N-terminal region and stimulating its catalytic activity, especially toward phosphorylated substrates [36]. It is possible that the mode of the activation of the enzyme differs depending on the activation of different cellular receptors and/or different subcellular localization of the enzyme. A recent study suggested the presence of nuclear localization sequence within the C2 domain which is important for the nuclear localization of PI3K-C2a, although the difference between the localization of fusion proteins and endogenous enzyme indicated that in addition to the nuclear localization sequence, other sequences are necessary for the distinct localization within nuclear speckles [34]. A remarkable finding is that the time course of the increase in the activity of PI3K-C2h closely correlates with the changes in the level of nuclear DAG observed previously in aphidicolin-synchronized HL-60 cells. The peak of the increase in nuclear DAG levels and the activity of PIPLC were measured at 8 – 9 h after the release from G1/S block and both flow cytometric and mitotic index analyses indicated that this peak of DAG coincides with the G2/M transition [5]. In the model of rat liver nuclei, previous

70

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

studies showed that the time course of nuclear PLC activation, DAG generation and PKC translocation corresponds well with that of the nuclear PI3K-C2h activation and the formation of PtdIns(3)P; the maximal increase in the activity of both PI-PLC and PI3K-C2h occurred at 20– 26 h after partial hepatectomy which is the peak of proliferation of liver cells during compensatory growth [18]. The mode of activation of both enzymes in the nuclei, their downstream targets and the precise role in the control of the cell cycle are not well understood. The mechanism of activation of PI3KC2h in the nuclei of either synchronized HL-60 cells or rat liver nuclei is not unique for the nuclear activation of the enzyme as the similar pattern was observed in the plasma membranes in response to fibrinogen [24]. Similarly, a possible role of Vav adaptor, which is known to operate at the plasma membrane, was suggested in the activation of a nuclear class IA PI3-kinases in HL-60 cells as an association of Vav adaptor molecule to PI3-kinase was described in the nuclei of ATRA-differentiated HL-60 cells [37]. Recently, a novel GTPase specific for nuclei, PI3K enchancer (PIKE), was described to mediate the nuclear activation of PI3kinase and G1 cell cycle arrest in PC12 cells [38]. Even less is known about the putative targets of activated nuclear PI3kinases. The increase in the level of nuclear DAG in aphidicolin-synchronized HL-60 cells is clearly associated with the activation of nuclear PKChII, which phosphorylates lamin B and thus participates in the nuclear lamina disassembly and entry into mitosis [5 –7]. The only product of the activation of PI3K-C2h in the nuclei of either synchronized HL-60 cells or rat liver nuclei is PtdIns(3)P [18]. As potential targets for PtdIns(3)P so far include mostly FYVEcontaining proteins [39] and PX domains [40,41], which are involved in different vesicle trafficking events, such as secretion and vacuole targeting, it is not clear what the possible mechanism is which links the production of nuclear PtdIns(3)P with the progression of cells through the cell cycle and/or cell differentiation [18,19]. However, the present study, together with recently published results [18,19], suggest that there are two ways of activation of nuclear PI3K-C2h one is by calpain-mediated proteolytic activation of the enzyme, when cells progress through the cell cycle [18], while the other is by tyrosine phosphorylation of the enzyme, when the cells are induced to differentiate [19]. In summary, the data presented in this report suggest that in the synchronized HL-60 cells the activity of PI3K-C2h in the nuclei and nuclear envelopes is cell-cycle regulated and rises to a peak corresponding to G2/M phase of the cell cycle. The increase in the activity of nuclear PI3K-C2h and the formation of PtdIns(3)P is sensitive to the presence of calpain inhibitor and could be mimicked by exogenous calpain, further suggesting that the activation of PI3K-C2h may be a calpain-mediated effect. Additional studies are necessary to define the downstream targets of the nuclear PI3K-C2h/PtdIns(3)P and further assessing its role in the regulation of the cell cycle progression.

Acknowledgements We thank Mr. Rene Lui for editing the manuscript before submission and HSM-informatika for editing the figures before submission. This work was supported by the Ministry of Science of the Republic of Croatia and by a Fogarty International Research Collaboration Award (to H.B.), and by the Italian Association for Cancer Research (to S.V.).

References [1] N. Divecha, H. Banfic, R.F. Irvine, The polyphosphoinositide cycle exists in the nuclei of Swiss 3T3 cells under the control of a receptor (for IGF-I) in the plasma membrane, and stimulation of the cycle increases nuclear diacylglycerol and apparently induces translocation of protein kinase C to the nucleus, EMBO J. 10 (1991) 3207 – 3214. [2] H. Banfic, M. Zizak, N. Divecha, R.F. Irvine, Nuclear diacylglycerol is increased during cell proliferation in vivo, Biochem. J. 290 (1993) 633 – 636. [3] J.D. York, P.W. Majerus, Nuclear phosphatidylinositols decrease during S-phase of the cell cycle in HeLa cells, J. Biol. Chem. 269 (1994) 7847 – 7850. [4] J.H. Clarke, A.J. Letcher, C.S. D’santos, J.R. Halstead, R.F. Irvine, N. Divecha, Inositol lipids are regulated during cell cycle progression in the nuclei of murine erythroleukaemia cells, Biochem. J. 357 (2001) 905 – 910. [5] B. Sun, N.R. Murray, A.P. Fields, A role for nuclear phosphatidylinositol-specific phospholipase C in the G2/M phase transition, J. Biol. Chem. 272 (1997) 26313 – 26317. [6] L.J. Thompson, A.P. Fields, Beta II protein kinase C is required for the G2/M phase transition of cell cycle, J. Biol. Chem. 271 (1996) 15045 – 15053. [7] V.L. Goss, B.A. Hocevar, L.J. Thompson, C.A. Stratton, D.J. Burns, A.P. Fields, Identification of nuclear beta II protein kinase C as a mitotic lamin kinase, J. Biol. Chem. 269 (1994) 19074 – 19080. [8] B. Vanhaesebroeck, M.D. Waterfield, Signalling by distinct classes of phosphoinositide 3-kinases, Exp. Cell Res. 253 (1999) 239 – 254. [9] A.M. Martelli, P. Borgatti, R. Bortul, M. Manfredini, L. Massari, S. Capitani, L.M. Neri, Phosphatidylinositol 3-kinase translocates to the nucleus of osteoblast-like MC3T3-E1 cells in response to insulin-like growth factor I and platelet-derived growth factor but not to the proapoptotic cytokine tumor necrosis factor alpha, J. Bone Miner. Res. 9 (2000) 1716 – 1730. [10] K. Tanaka, K. Horiguchi, T. Yoshida, M. Takeda, H. Fujisawa, K. Takeuchi, M. Umeda, S. Kato, S. Ihara, S. Nagata, Y. Fukui, Evidence that a phosphatidylinositol 3,4,5-trisphosphate-binding protein can function in nucleus, J. Biol. Chem. 274 (1999) 3919 – 3922. [11] A. Metjian, R.L. Roll, A.D. Ma, C.S. Abrams, Agonists cause nuclear translocation of phosphatidylinositol 3-kinase gamma. A betagammadependent pathway that requires the p110gamma amino terminus, J. Biol. Chem. 274 (1999) 27943 – 27947. [12] D. Bacqueville, P. Deleris, C. Mendre, M.T. Pieraggi, H. Chap, G. Guillon, B. Perret, M. Breton-Douillon, Characterization of a G protein-activated phosphoinositide 3-kinase in vascular smooth muscle cell nuclei, J. Biol. Chem. 276 (2001) 22170 – 22176. [13] S. Volinia, R. Dhand, B. Vanhaesebroeck, L.K. MacDougall, R. Zvelebil, M.J. Zvelebil, J. Domin, C. Panaretou, M.D. Waterfield, A human phosphatidylinositol 3-kinase complex related to the yeast Vps34p – Vps15p protein sorting system, EMBO J. 14 (1995) 3339 – 3348. [14] J. Domin, F. Pages, S. Volinia, S.E. Rittenhouse, M.J. Zvelebil, R.C. Stein, M.D. Waterfield, Cloning of a human phosphoinositide 3-kin-

D. Visˇnjic´ et al. / Biochimica et Biophysica Acta 1631 (2003) 61–71

[15]

[16]

[17]

[18]

[19]

[20]

[21]

[22]

[23]

[24]

[25] [26]

ase with a C2 domain that displays reduced sensitivity to the inhibitor wortmannin, Biochem. J. 326 (1997) 139 – 147. R.A. Brown, L.K. Ho, S.J. Weber-Hall, J.M. Shipley, M.J. Fry, Identification and cDNA cloning of a novel mammalian C2 domain-containing phosphoinositide 3-kinase, HsC2-PI3K, Biochem. Biophys. Res. Commun. 233 (1997) 537 – 544. A. Arcaro, S. Volinia, M.J. Zvelebil, R. Stein, S.J. Watton, M.J. Layton, I. Gout, K. Ahmadi, J. Downward, M.D. Waterfield, Human phosphoinositide 3-kinase C2beta, the role of calcium and the C2 domain in enzyme activity, J. Biol. Chem. 273 (1998) 33082 – 33090. F. Ono, T. Nakagawa, S. Saito, Y. Owada, H. Sakagami, K. Goto, M. Suzuki, S. Matsuno, H. Kondo, A novel class II phosphoinositide 3-kinase predominantly expressed in the liver and its enhanced expression during liver regeneration, J. Biol. Chem. 273 (1998) 7731 – 7736. A. Sindic, A. Aleksandrova, A.P. Fields, S. Volinia, H. Banfic, Presence and activation of nuclear phosphoinositide 3-kinase C2beta during compensatory liver growth, J. Biol. Chem. 276 (2001) 17754 – 17761. D. Visnjic, V. Crljen, J. Curic, D. Batinic, S. Volinia, H. Banfic, The activation of nuclear phosphoinositide 3-kinase C2beta in all-transretinoic acid-differentiated HL-60 cells, FEBS Lett. 529 (2002) 268 – 274. A.P. Fields, G.R. Pettit, W.S. May, Phosphorylation of lamin B at the nuclear membrane by activated protein kinase C, J. Biol. Chem. 263 (1988) 8253 – 8260. T. Wakabayashi, M. Asano, C. Kurono, H. Kimura, Effects of extremely low concentrations of glutaraldehyde together with shortening of fixation time on the maintenance of various enzymes activities, J. Histochem. Cytochem. 23 (1975) 632 – 634. R.C. Nordlie, W.R. Arion, Liver microsomal glucose-6-phosphatase, inorganic pyrophosphatase, and pyrophosphate-glucose phosphotransferase. Effects of adrenalectomy and cortisone administration on activities assayed in the absence and presence of deoxycholate, J. Biol. Chem. 240 (1965) 3479 – 3484. J. Burnside, D.L. Schneider, Characterization of the membrane proteins of rat liver lysosomes. Composition, enzyme activities and turnover, Biochem. J. 204 (1982) 525 – 534. J. Zhang, H. Banfic, F. Straforini, L. Tosi, S. Volinia, S.E. Rittenhouse, A type II phosphoinositide 3-kinase is stimulated via activated integrin in platelets. A source of phosphatidylinositol 3-phosphate, J. Biol. Chem. 273 (1998) 14081 – 14084. U.K. Laemmli, Cleavage of structural proteins during the assembly of the head of bacteriophage T4, Nature (London) 227 (1970) 680 – 685. L.R. Vann, F.B.P. Wooding, R.F. Irvine, N. Divecha, Metabolism and possible compartmentalization of inozitol lipids in isolated rat-liver nuclei, Biochem. J. 327 (1997) 569 – 576.

71

[27] R.L. Mellgren, Q. Lu, Selective nuclear transport of mu-calpain, Biochem. Biophys. Res. Commun. 204 (1994) 544 – 550. [28] R.L. Mellgren, Proteolysis of nuclear proteins by mu-calpain and mcalpain, J. Biol. Chem. 266 (1991) 13920 – 13924. [29] M. Molinari, E. Carafoli, Calpain: a cytosolic proteinase active at the membranes, J. Membr. Biol. 156 (1997) 1 – 8. [30] S.J. Turner, J. Domin, M.D. Waterfield, S.G. Ward, J. Westwick, The CC chemokine monocyte chemotactic peptide-1 activates both the class I p85/p110 phosphatidylinositol 3-kinase and the class II PI3KC2alpha, J. Biol. Chem. 273 (1998) 25987 – 25995. [31] R.A. Brown, J. Domin, A. Arcaro, M.D. Waterfield, P.R. Shepherd, Insulin activates the alpha isoform of class II phosphoinositide 3kinase, J. Biol. Chem. 274 (1999) 14529 – 14532. [32] A. Arcaro, M.J. Zvelebil, C. Wallasch, A. Ullrich, M.D. Waterfield, J. Domin, Class II phosphoinositide 3-kinases are downstream targets of activated polypeptide growth factor receptors, Mol. Cell. Biol. 20 (2000) 3817 – 3830. [33] J. Domin, I. Gaidarov, M.E. Smith, J.H. Keen, M.D. Waterfield, The class II phosphoinositide 3-kinase PI3K-C2alpha is concentrated in the trans-Golgi network and present in clathrin-coated vesicles, J. Biol. Chem. 275 (2000) 11943 – 11950. [34] S.A. Didichenko, M. Thelen, Phosphatidylinositol 3-kinase C2alpha contains a nuclear localization sequence and associates with nuclear speckles, J. Biol. Chem. 276 (2001) 48135 – 48142. [35] M. Wheeler, J. Domin, Recruitment of the class II phosphoinositide 3kinase C2beta to the epidermal growth factor receptor: role of Grb2, Mol. Cell. Biol. 21 (2001) 6660 – 6667. [36] I. Gaidarov, M.E. Smith, J. Domin, J.H. Keen, The class II phosphoinositide 3-kinase C2alpha is activated by clathrin and regulates clathrin-mediated membrane trafficking, Mol. Cell 7 (2001) 443 – 449. [37] V. Bertagnolo, M. Marchisio, S. Volinia, E. Caramelli, S. Capitani, Nuclear association of tyrosine-phosphorylated Vav to phospholipase C-gamma1 and phosphoinositide 3-kinase during granulocytic differentiation of HL-60 cells, FEBS Lett. 441 (1998) 480 – 484. [38] K. Ye, K.J. Hurt, F.Y. Wu, M. Fang, H.R. Luo, J.J. Hong, S. Blackshaw, C.D. Ferris, S.H. Snyder, Pike. A nuclear GTPase that enhances PI3kinase activity and is regulated by protein 4.1N, Cell 103 (2000) 919 – 930. [39] B. Vanhaesebroeck, S.J. Leevers, K. Ahmadi, J. Timms, R. Katso, P.C. Driscoll, R. Woscholski, P.J. Parker, M.D. Waterfield, Synthesis and function of 3-phosphorylated inositol lipids, Ann. Rev. Biochem. 70 (2001) 535 – 602. [40] A. Simonsen, H. Stenmark, PX domains: attracted by phosphoinositides, Nat. Cell Biol. 3 (2001) E179 – E182. [41] M.J. Wishart, G.S. Taylor, J.E. Dixon, Phoxy lipids: revealing PX domains as phosphoinositide binding modules, Cell 105 (2001) 817 – 820.