Advertisement

Phosphatidylinositol 3-Kinase C2α Contains a Nuclear Localization Sequence and Associates with Nuclear Speckles*

  • Svetlana A. Didichenko
    Affiliations
    Institute for Research in Biomedicine, Via Vincenzo Vela 6, CH 6500 Bellinzona, Switzerland
    Search for articles by this author
  • Marcus Thelen
    Correspondence
    To whom correspondence should be addressed. Tel.: 41 91 820 0317; Fax: 41 91 820 0305
    Affiliations
    Institute for Research in Biomedicine, Via Vincenzo Vela 6, CH 6500 Bellinzona, Switzerland
    Search for articles by this author
  • Author Footnotes
    * This work was supported in part by the Helmut Horten Foundation.The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
Open AccessPublished:December 21, 2001DOI:https://doi.org/10.1074/jbc.M104610200
      Phosphoinositide 3-kinase C2α (PI3K-C2α) belongs to the class II phosphatidylinositol 3-kinases, which are defined by their in vitro usage of phosphatidylinositol and phosphatidylinositol 4-phosphate as substrates. All type II phosphatidylinositol 3-kinases contain at their C terminus a C2-like domain. Here we demonstrate that Homo sapiens phosphoinositide 3-kinase C2α (HsPI3K-C2α) has dual cellular localization present in the cytoplasm and in the nucleus. A distinct nuclear localization signal sequence was identified by expressing HsPI3K-C2α-green fluorescent protein fusion proteins in HeLa cells. The nuclear localization signal was mapped to a stretch of 11 amino acids (KRKTKISRKTR) located within C2-like domain of the kinase. In the cytoplasm and the nucleus HsPI3K-C2α associates with macromolecular complexes that are resistant to detergent extraction. Indirect immunofluorescence reveals that in the nucleus HsPI3K-C2α is enriched at distinct subnuclear domains known as nuclear speckles, which contain pre-mRNA processing factors and are functionally connected to RNA metabolism. Phosphorylation of HsPI3K-C2α is induced by inhibition of RNA polymerase II-dependent transcription and coincides with enlargement and rounding up of the nuclear speckles. The results suggest that phosphorylation of HsPI3K-C2α is inversely linked to mRNA transcription and supports the importance of phosphoinositides for nuclear activity.
      PI 3-kinase
      phosphatidylinositol 3-kinase
      HsPI3K-C2α
      Homo sapiens phosphatidylinositol 3-kinase C2α
      PtdIns
      phosphatidylinositol
      PtdIns (4
      5)P2, PtdIns 4,5-bisphosphate
      NLS
      nuclear localization signal
      GFP
      green fluorescent protein
      PBS
      phosphate-buffered saline
      DTT
      dithiothreitol
      PNS
      postnuclear supernatant
      Pipes
      1,4-piperazinediethanesulfonic acid
      snRNP
      small nuclear ribonucleoprotein
      Phosphatidylinositol 3-kinases (PI 3-kinases)1 have emerged as important constituents of cellular pathways regulating the remodeling of the cytoskeleton, the trafficking of intracellular organelles, and cell growth and survival (
      • Fruman D.A.
      • Meyers R.E.
      • Cantley L.C.
      ). Type I PI 3-kinases are primarily involved in receptor-mediated signal transduction and preferably use PtdIns (4,5)P2 as the substrate in vivo but also phosphorylate phosphatidylinositol and phosphatidylinositol 4-phosphate at the 3-OH position of the inositol ring. The type III PI 3-kinases exclusively phosphorylate PtdIns and are essential for vesicular trafficking. Both type I and type III PI 3-kinases are heterodimeric enzymes consisting of a catalytic and a regulatory subunit. By contrast, the type II PI 3-kinases are monomeric enzymes that use PtdIns and PtdIns 4-phosphate as substrates. Their function in signal transduction is poorly understood. Few studies provide evidence for a role of type II kinases in chemokine and growth factor receptor-mediated cell activation (
      • Turner S.J.
      • Domin J.
      • Waterfield M.D.
      • Ward S.G.
      • Westwick J.
      ,
      • Arcaro A.
      • Volinia S.
      • Zvelebil M.J.
      • Stein R.
      • Watton S.J.
      • Layton M.J.
      • Gout I.
      • Ahmadi K.
      • Downward J.
      • Waterfield M.D.
      ,
      • Arcaro A.
      • Zvelebil M.J.
      • Wallasch C.
      • Ullrich A.
      • Waterfield M.D.
      • Domin J.
      ,
      • Brown R.A.
      • Domin J.
      • Arcaro A.
      • Waterfield M.D.
      • Shepherd P.R.
      ,
      • Sindic A.
      • Aleksandrova A.
      • Fields A.P.
      • Volinia S.
      • Banfic H.
      ) and in integrin-mediated platelet activation (
      • Zhang J.
      • Banfic H.
      • Straforini F.
      • Tosi L.
      • Volinia S.
      • Rittenhouse S.E.
      ).
      A common structural characteristic of type II PI 3-kinases is a C2-like domain at the C terminus. The C2 domains of many proteins mediate calcium-dependent phospholipid binding (
      • Newton A.C.
      ). However, type II PI 3-kinases lack a critical aspartate residue in their C2-like domain and the requirement of calcium for membrane binding and catalytic activity remains controversial (
      • Arcaro A.
      • Zvelebil M.J.
      • Wallasch C.
      • Ullrich A.
      • Waterfield M.D.
      • Domin J.
      ,
      • Domin J.
      • Waterfield M.D.
      ). Three human type II PI 3-kinases have been characterized; they are HsPI3K-C2α, -β, and -γ. They share a similar structure containing an unique N terminus, a catalytic domain, a PX domain, and a C2-like domain at the C terminus (
      • Wymann M.P.
      • Pirola L.
      ). Although HsPI3K-C2α and HsPI3K-C2β share wide tissue distribution, HsPI3K-C2γ expression is restricted to hepatocytes and is enhanced during liver regeneration (
      • Misawa H.
      • Ohtsubo M.
      • Copeland N.G.
      • Gilbert D.J.
      • Jenkins N.A.
      • Yoshimura A.
      ,
      • Ono F.
      • Nakagawa T.
      • Saito S.
      • Owada Y.
      • Sakagami H.
      • Goto K.
      • Suzuki M.
      • Matsuno S.
      • Kondo H.
      ). Both HsPI3K-C2α and HsPI3K-C2β are implicated in signaling downstream of epidermal growth factor and platelet-derived growth factor receptors (
      • Arcaro A.
      • Zvelebil M.J.
      • Wallasch C.
      • Ullrich A.
      • Waterfield M.D.
      • Domin J.
      ). Signaling through insulin receptor or chemokine receptors induces activation of HsPI3K-C2α (
      • Turner S.J.
      • Domin J.
      • Waterfield M.D.
      • Ward S.G.
      • Westwick J.
      ,
      • Brown R.A.
      • Domin J.
      • Arcaro A.
      • Waterfield M.D.
      • Shepherd P.R.
      ). Recent reports demonstrate that HsPI3K-C2α is concentrated in trans-Golgi network and is present in clathrin-coated pits (
      • Domin J.
      • Gaidarov I.
      • Smith M.E.
      • Keen J.H.
      • Waterfield M.D.
      ), whereas PI3K-C2β was found in the nuclei of rat liver cells (
      • Sindic A.
      • Aleksandrova A.
      • Fields A.P.
      • Volinia S.
      • Banfic H.
      ).
      Phosphoinositide signaling in the nucleus is regulated independently from plasma membrane phosophoinositide pathways. Some nuclear phosphoinositides and their metabolizing enzymes are not extracted with non-ionic detergents, which indicates that they are not associated with membrane structures (
      • Payrastre B.
      • Nievers M.
      • Boonstra J.
      • Breton M.
      • Verkleij A.J.
      • van Bergen en Henegouwen P.M.
      ,
      • Boronenkov I.V.
      • Loijens J.C.
      • Umeda M.
      • Anderson R.A.
      ). Two PtsIns 4-phosphate 5-kinases (Iα and IIα) are reported to localize to nuclear speckles together with their product PI (4,5)P2 (
      • Boronenkov I.V.
      • Loijens J.C.
      • Umeda M.
      • Anderson R.A.
      ). By electron microscopy analysis, speckles were found to consist of two morphologically and functionally distinct domains. The larger and denser regions seen by fluorescence microscopy correspond to interchromatin granule clusters, which are not active in transcription. The more diffusely distributed splicing factors and the regions of the periphery of the interchromatin granule clusters correspond to perichromatin fibrils, where transcription takes place (
      • Fu X.D.
      • Maniatis T.
      ,
      • Spector D.L.
      • Fu X.D.
      • Maniatis T.
      ,
      • Spector D.L.
      ). The composition of nuclear speckles is highly dynamic because pre-mRNA transcribing and processing factors move rapidly in and out and are concentrated by transient associations with functionally related components (
      • Lewis J.D.
      • Tollervey D.
      ). Thus, the overall morphological appearance of nuclear speckles reflects the transcriptional activity of the cell. (
      • Phair R.D.
      • Misteli T.
      ). In response to inhibition of RNA polymerase II-dependent transcription, speckles become larger, round up, and lose their irregular shape (
      • Sinclair G.D.
      • Brasch K.
      ,
      • O'Keefe R.T.
      • Mayeda A.
      • Sadowski C.L.
      • Krainer A.R.
      • Spector D.L.
      ,
      • Spector D.L.
      • Schrier W.H.
      • Busch H.
      ). Furthermore, it has been proposed that cycles of phosphorylation and dephosphorylation control the subnuclear distribution of splicing factors, thereby regulating their association with the speckles (
      • Misteli T.
      • Spector D.L.
      ,
      • Gui J.F.
      • Lane W.S.
      • Fu X.D.
      ).
      We show here that HsPI3K-C2α is present in the cytoplasm and in the nucleus. In many cases translocation of proteins into the nucleus is conferred by distinct import signals. The best characterized are nuclear localization signals (NLS) consisting of one or more clusters of basic amino acids (
      • Dingwall C.
      • Laskey R.A.
      ). A signal sequence for nuclear localization of HsPI3K-C2α was mapped to a short stretch of highly basic amino acids localized within the C2 domain of the kinase, a domain that is known to target a variety of proteins to the plasma membrane. In the nucleus the HsPI3K-C2α concentrates at nuclear speckles together with pre-mRNA-processing factors.

      EXPERIMENTAL PROCEDURES

      Antibodies against PI3K-C2 were raised in rabbits immunized with chimeric proteins containing glutathione S-transferase fused either to amino acids 62–131 of murine p170 (GenBankTMaccession number U55772) (
      • Virbasius J.V.
      • Guilherme A.
      • Czech M.P.
      ) or to the N terminus (amino acids 1–134) of HsPI3K-C2α (GenBankTM accession number Y13367) (
      • Domin J.
      • Pages F.
      • Volinia S.
      • Rittenhouse S.E.
      • Zvelebil M.J.
      • Stein R.C.
      • Waterfield M.D.
      ). Immune sera were purified by affinity chromatography using the corresponding domains of p170 (antibody AXIX) and HsPI3K-C2α (antibody AXXIII) immobilized onN-hydroxysuccinimide-Sepharose (Amersham Pharmacia Biotech). Briefly, isolated glutathione S-transferase fusion proteins were proteolysed with thrombin, and the cleaved products were further purified by reverse phase high pressure liquid chromatography. Anti-lamin goat polyclonal IgG was purchased from Santa Cruz Biotechnology, anti-green fluorescent protein (GFP) rabbit polyclonal antibody was from CLONTECH, and horseradish peroxidase-conjugated goat anti-rabbit was from Bio-Rad. Fluorescent dye-conjugated secondary antibodies were from Jackson ImmunoResearch Laboratories. Human autoimmune Sm serum and anti-CAP monoclonal antibodies were kindly provided by Dr. A. Rosen (Baltimore, MD) and Dr. R. Luhrmann (Göttingen, Federal Republic of Germany), respectively.

      Plasmids

      The cDNA encoding human HsPI3K-C2α (
      • Domin J.
      • Pages F.
      • Volinia S.
      • Rittenhouse S.E.
      • Zvelebil M.J.
      • Stein R.C.
      • Waterfield M.D.
      ) was a kindly provided by Dr. J. Domin. DNA fragments corresponding to subdomains of HsPI3K-C2α were generated by the PCR using specific primers. For transient expression the fragments were cloned into the eukaryotic expression vector pEGFP (CLONTECH). Domains corresponding to the N terminus of HsPI3K-C2α were fused upstream of GFP (vector pEGFP-N1), whereas domains corresponding to the C terminus of HsPI3K-C2α were fused downstream of GFP (vector pEGFP-C1). Parts of the C2-like domain of HsPI3K-C2α were fused to the C terminus of red fluorescent protein (vector pDsRed1-C1,CLONTECH).

      Cell Culture and Transient Expressions

      HeLa (ATCC) cells were cultured at 37 °C in Dulbecco’s modified Eagle’s minimal essential medium (Life Technologies, Inc.) supplemented with 10% fetal calf serum and antibiotics. Treatment of HeLa cells with 5 μg/ml actinomycin D, 50 μg/ml cycloheximide, or 40 μg/ml α-amanitin (all from Sigma) was carried out for 5 h at 37 °C. For transfection, subconfluent cell cultures were trypsinized, washed, and resuspended at 2.5 × 106 cells/ml in Dulbecco’s modified Eagle’s medium containing 10% fetal calf serum and 15 mm HEPES (pH 7.5). To 200 μl of cell suspension 50 μl of DNA mixture (200 mm NaCl, 10 μg of plasmid DNA, and 30 μg of carrier DNA (salmon sperm, Stratagene)) was added. After electroporation (960 μF, 240 V) the cells were cultured for 16–48 h before analysis.

      Subcellular Fractionation

      HeLa cells (∼1–5 × 107 cells) were treated with cytochalasin B (10 μg/ml) in culture medium for 30 min, trypsinized, washed 2 times with PBS, and resuspended in 1 ml of ice-cold hypotonic lysis buffer (10 mm Tris-HCl (pH 7.5), 10 mm NaCl, 1.5 mm MgCl2, 1 mm EGTA, 1 mm EDTA, 1 mm DTT, 10 μg/ml cytochalasin B, 40 mm NaF, 0.5 mm sodium orthovanadate, 40 mm β-glycerophosphate, 5 mm sodium pyrophosphate, and protease inhibitors (Complete, Roche Molecular Biochemicals)). Cells were allowed to swell on ice for 10 min and then passed several times through a 27-gauge syringe needle. The cell homogenate was layered onto 300 μl of 30% sucrose (w/v) in lysis buffer and spun at 2,000 × g for 10 min. The postnuclear supernatant (PNS) containing membranes and cytosol was collected and further processed. The pellet consisting of nuclei and few unbroken cells was resuspended in lysis buffer, homogenized, divided into four equal parts, and centrifuged as before. After an additional wash with lysis buffer, nuclear pellets were resuspended either in lysis buffer and left on ice for 15 min or in lysis buffer containing 1% Triton X-100 and placed for 15 min on ice or in high salt buffer (20 mm HEPES (pH 7.5), 0.4 m NaCl, 1.5 mm MgCl2, 1 mm EGTA, 1 mm EDTA, 1 mm DTT) and incubated on ice for 30 min or in 10 mm Tris-Cl (pH 7.5), 10 mm NaCl, 1.5 mm MgCl2 containing RNase free DNase I (20 units/ml, Roche Molecular Biochemicals) and incubated for 20 min on ice. The aliquots were then centrifuged at 2,000 × g for 10 min. Proteins in the supernatants and pellets were precipitated with 10% trichloroacetic acid, washed with cold acetone, and finally dissolved in SDS sample buffer (60 mm Tris-HCl (pH 6.8), 10% glycerol, 2% SDS, 100 mm DTT, 0.01% bromphenol blue), boiled for 5 min, and analyzed on SDS-polyacrylamide gels.
      Nuclear envelopes were separated from the nuclear extracts as follows. Nuclei were resuspended in buffer (10 mm Tris-HCl (pH 8), 10 mm NaCl, 3 mm MgCl2, 0.1 mm EDTA, 8 units/ml RNasin (Promega), 40 mmNaF, 0.5 mm NaVO4, 40 mmβ-glycerophosphate, 5 mm sodium pyrophosphate, and protease inhibitors (Complete, Roche Molecular Biochemicals)) and disrupted by sonication using a micro-tip (two 15-s pulses at 50 W). The homogenate was centrifuged for 30 min at 10,000 ×g to pellet nuclear envelopes, and the supernatant (nuclear extract) was further separated by high speed centrifugation (15 min, 400,000 × g) into a soluble fraction and a pellet.
      The PNS was fractionated into cytosol and a high speed pellet consisting of membranes, cytoskeleton, and ribosomal structures as follows. PNS was passed three times through a 27-gauge needle and cleared at 5,000 g for 15 min, and the resulting supernatant was centrifuged at 400,000 × g for 10 min at 4 °C.
      For RNase treatment high speed pellets prepared from PNS and nuclear extract were resuspended in RNase digestion buffer (10 mmTris-HCl (pH 8.0), 1% 2-mercaptoethanol) and incubated without or with RNases A (1 mg/ml) and T1 (50 units/ml) at 37 °C for 30 min. After treatment the samples were fractionated into supernatant and pellet at 400,000 × g for 15 min.

      Gel Electrophoresis, Immunoprecipitation, and Western Blot Analysis

      Proteins were separated on 8% or 6% SDS-polyacrylamide gels prepared from a acrylamide stock (33.5% acrylamide, 0.3% bisacrylamide) and blotted onto Immobilon-P (Millipore). Membranes were blocked with 5% nonfat milk in Tris-buffered saline containing 0.1% Triton X-100 and probed with specific antibodies. Immunoreactive bands were decorated with horseradish peroxidase-labeled antibodies or protein G-conjugated with horseradish peroxidase (Zymed Laboratories Inc.) and visualized by enhanced chemiluminescence (Pierce).
      For immunoprecipitation, cells were washed twice in PBS and lysed in buffer (1% Nonidet P-40, 150 mm NaCl, 50 mm Tris-HCl (pH 7.5), 1.5 mm MgCl2, 1 mm EDTA) supplemented with phosphatase inhibitors (40 mm NaF, 0.5 mm sodium orthovanadate, 40 mm β-glycerophosphate, 5 mm sodium pyrophosphate) and protease inhibitors (Complete, Roche Molecular Biochemicals). Cell homogenates were centrifuged at 13,000 ×g for 10 min, and supernatants were precleared with GammaBind G-Sepharose (Amersham Pharmacia Biotech) for 15 min. Immunoprecipitation of HsPI3K-C2α with antibody AXXIII was carried out at 4 °C for 1–2 h. Immune complexes were bound to GammaBind G-Sepharose for 30 min, collected by centrifugation, and washed two times in lysis buffer, once in 10 mm Tris-HCl (pH 8), 0.5m NaCl, 0.5% Nonidet P-40, 0.5% deoxycholate, 0.05% SDS, then in 10 mm Tris-HCl (pH 8), once in 10 mmTris-HCl (pH 8), 150 mm NaCl, 0.5% Nonidet P-40, 0.5% deoxycholate, 0.05% SDS, and finally in 10 mm Tris-HCl (pH 8), 0.05% SDS. For λ-phosphatase treatment immunoprecipitates were additionally washed twice in phosphatase buffer (50 mmTris-HCl (pH 7.5), 2 mm MnCl2, 0.1 mm EDTA, 5 mm DTT, 0.01% Brij 35) and resuspended in 50 μl of the same buffer. After warming up at 30 °C for 3 min, 50 units of λ-phosphatase (New England Biolabs) was added, and the samples were incubated at 30 °C for 40 min.

      Kinase Assay

      HsPI3K-C2α was immunoprecipitated as described in the previous section except that the washing steps were omitted and, instead, immune complexes bound to GammaBind were washed three time with lysis buffer and then treated with buffer or λ-phosphatase as mentioned above. After phosphatase treatment, immunoprecipitates were washed twice with 0.5 m LiCl, 50 mm Tris-HCl (pH 8.0), once with 150 mm NaCl, 10 mm Tris-HCl (pH 7.5), 1 mm EDTA, and once with 20 mm Hepes (pH 7.3), 1 mm DTT, 5 mm MgCl2. The resulting immunoprecipitates were resuspended in kinase assay buffer (50 mm Tris-HCl (pH 7.6), 50 mm NaCl, 20 mm glycerophosphate, 0.1 mm sodium vanadate, 10 mm NaF). PI 3-kinase activity was determined in of 50 μl of kinase assay buffer containing 10 μg each PtdIns and phosphatidylserine, 0.1% sodium cholate, 0.25 mm EDTA, 1 mm dithiothreitol, 5 mmMgCl2, and 0.1 mm ATP. Samples were warmed up for 10 min at 30 °C, and reactions were initiated by the addition of MgCl2 and ATP (containing 5–10 μCi of [γ-32P]ATP (Amersham Pharmacia Biotech)). Reactions were terminated after 20 min with 100 μl of 1 m HCl, and the lipids were extracted with 200 μl of chloroform:methanol (1:1). The aqueous phase was washed twice with 80 μl of 1 mHCl:methanol (1:1), and the lipids were dried and analyzed by thin layer chromatography as described (
      • Didichenko S.A.
      • Tilton B.
      • Hemmings B.A.
      • Ballmer-Hofer K.
      • Thelen M.
      ).

      Immunofluorescence

      HeLa cells, grown on glass coverslips, were permeabilized with 1% Triton X-100, 2 mm EGTA, 5 mm Pipes (pH 6.7) for 1 min at room temperature and immediately fixed in cold (−70 °C) methanol for 10 min or in 0.4% paraformaldehyde for 20 min at room temperature. After several washes in PBS the coverslips were blocked in PGB (PBS containing 10% goat serum (Sigma) and 0.5% bovine serum albumin (Sigma)) for 15 min followed by the incubation with the primary antibody for 1 h in a humidified chamber. Antibodies were diluted in PGB and used at the following concentrations: affinity-purified anti-HsPI3K-C2α (5 μg/ml), anti-Sm serum (1:1000 dilution). The coverslips were then washed several times with PBS and incubated again with PGB for 10 min. Fluorescence dye-conjugated secondary antibodies (1 μg/ml) were added for 1 h in PGB. Coverslips were then washed extensively with PBS and once with water and mounted in polyvinyl alcohol (Gelvatol, Sigma) supplemented with 1% 1,4-diazabicyclo[2,2,2]octane (Sigma).

      DISCUSSION

      In this study we demonstrate that in resting cells the type II PI 3-kinase HsPI3K-C2α resides in the cytoplasm and in the nucleus. Recently it has been shown that the structurally related PI3K-C2β is associated with membrane-depleted nuclei of rat liver cells (
      • Sindic A.
      • Aleksandrova A.
      • Fields A.P.
      • Volinia S.
      • Banfic H.
      ). These observations reveal a remarkable difference with the localization of type I PI 3-kinases, which are generally considered cytosolic proteins (
      • Fruman D.A.
      • Meyers R.E.
      • Cantley L.C.
      ,
      • Thelen M.
      • Wymann M.P.
      • Langen H.
      ,
      • Stephens L.E.
      • Eguinoa A.
      • Erdjument-Bromage H.
      • Lui M.
      • Cooke F.
      • Coadwell J.
      • Smrcka A.
      • Thelen M.
      • Cadwallader K.
      • Tempst P.
      • Hawkins P.T.
      ). Few reports provide evidence for type IA PI 3-kinase activity in the nuclei of stimulated cells such as human osteosarcoma (
      • Zini N.
      • Ognibene A.
      • Bavelloni A.
      • Santi S.
      • Sabatelli P.
      • Baldini N.
      • Scotlandi K.
      • Serra M.
      • Maraldi N.M.
      ) and rat liver cells (
      • Lu P.J.
      • Hsu A.L.
      • Wang D.S.
      • Yan H.Y.
      • Yin H.L.
      • Chen C.S.
      ). The catalytic subunit of type IB PI 3-kinase γ has recently been demonstrated to translocate to the nuclei of HepG2 cells after stimulation with serum (
      • Metjian A.
      • Roll R.L.
      • Ma A.D.
      • Abrams C.S.
      ). However, the mechanism of translocation and activation of PI3-kinases in the nucleus is poorly understood.
      Our data provide evidence that HsPI3K-C2α concentrates in the interchromatin granule clusters (nuclear speckles). Nuclear speckles are known to contain factors involved in the transcription and processing of pre-mRNA including RNA polymerase II, snRNPs, and non-snRNP splicing factors (
      • Misteli T.
      • Spector D.L.
      ). Splicing factors associated with speckles are resistant to extraction with non-ionic detergent, high salt, or treatment with DNase I, suggesting that speckles represent nuclear compartments that are possibly connected with the nuclear scaffold (
      • Fey E.G.
      • Krochmalnic G.
      • Penman S.
      ,
      • Bisotto S.
      • Lauriault P.
      • Duval M.
      • Vincent M.
      ). We show that HsPI3K-C2α exhibits all these properties common to splicing factors. The overall morphological appearance of nuclear speckles is indicative for the transcriptional activity of the cell. Upon inhibition of RNA polymerase II the speckles lose their irregular shape and collapse into round and larger clusters (
      • Fey E.G.
      • Krochmalnic G.
      • Penman S.
      ,
      • Carmo-Fonseca M.
      • Pepperkok R.
      • Carvalho M.T.
      • Lamond A.I.
      ). HsPI3K-C2α was found to redistribute identically with other speckle components (Sm-antigenes) in response to inhibition of RNA polymerase II by α-amanitin (Fig. 3). The observation suggests that HsPI3K-C2α directly interacts with some components of nuclear speckles. It is conceivable that the kinase is in a complex with some type of RNA because treatment with RNases leads to its solubilization (Fig. 2 C). A likely candidate for the interaction is poly(A)+ mRNA, since HsPI3K-C2α was found in preparations of RNPs containing poly (A)+ mRNA that were purified by affinity chromatography on oligo(dT)-cellulose.
      S. A. Didichenko and M. Thelen, our unpublished results.
      We could not detect HsPI3K-C2α in snRNP preparations, and in vitro splicing assays appeared to be insensitive to the PI 3-kinase inhibitor wortmannin.
      R. Lurhman, personal communication.
      Therefore, the direct interaction of HsPI3K-C2α with snRNP splicing factors is less probable.
      The reversible phosphorylation of splicing factors on serine/threonine residues is known to affect their localization in nuclear speckles and regulate their activity (
      • Misteli T.
      • Spector D.L.
      ,
      • Gui J.F.
      • Lane W.S.
      • Fu X.D.
      ,
      • Colwill K.
      • Pawson T.
      • Andrews B.
      • Prasad J.
      • Manley J.L.
      • Bell J.C.
      • Duncan P.I.
      ). We found that inhibition of transcription, which causes the collapse of nuclear speckles, is accompanied by the phosphorylation of HsPI3K-C2α. It is, therefore, plausible that the state of phosphorylation is critical for the subnuclear localization of HsPI3K-C2α. In agreement with the continued association with nuclear speckles, we did not obtain evidence that phosphorylation of HsPI3K-C2α causes redistribution of the kinase between the nucleus and the cytosol. In the cytosol phosphorylation of HsPI3K-C2α did not cause a detectable change of the immunostaining pattern. Preliminary data suggest that the phosphorylation of HsPI3K-C2α induced by inhibition of transcription occurs at serine residues. Protein kinases SRPK-1 (
      • Gui J.F.
      • Lane W.S.
      • Fu X.D.
      ), CLK/STY (
      • Colwill K.
      • Pawson T.
      • Andrews B.
      • Prasad J.
      • Manley J.L.
      • Bell J.C.
      • Duncan P.I.
      ), and casein kinase Iα (
      • Gross S.D.
      • Loijens J.C.
      • Anderson R.A.
      ) localize to nuclear speckles and phosphorylate non-snRNP-splicing factors (SR proteins). Nonetheless, it is unlikely that HsPI3K-C2α is phosphorylated by either kinase since HsPI3K-C2α lacks the serine/arginine-rich consensus essential for phosphorylation by SRPK-1 and CLK/STY and also does not contain an appropriate consensus sequence for casein kinase Iα. Identification of the kinase that phosphorylates HsPI3K-C2α may provide an insight on the regulation of the HsPI3K-C2α.
      We found a similar subcellular distribution of endogenous HsPI3K-C2α in cells from different tissues including human primary dendritic cells, mesenchymal cells (HeLa), pancreatic carcinomas (Aspc-1, T3M4), and epithelial tumor cells (A549) as well as murine macrophages (J7741.A) and fibroblasts (NIH3T3). Our immunofluorescence data do not provide evidence that HsPI3K-C2α associates with the trans-Golgi network. This contrasts somehow with a previous report by Dominet al. (
      • Domin J.
      • Gaidarov I.
      • Smith M.E.
      • Keen J.H.
      • Waterfield M.D.
      ) who show by indirect immunofluorescence that in HEK293 cells HsPI3K-C2α co-localizes with γ-adaptin. Accessibility of the HsPI3K-C2α for different antibodies and differences in the fixation protocols used during immunofluorescence staining could be the reason for the discrepancy. However, we observed a similar nuclear localization of HsPI3K-C2α in non-fixed HeLa cells, which were just permeabilized with Triton X-100 before staining (not shown). This particular method has been applied to reveal the dynamic distribution of pre-mRNA-splicing factors localized to nuclear speckles (
      • Misteli T.
      • Caceres J.F.
      • Spector D.L.
      ).
      In the cytoplasm HsPI3K-C2α appears to be tightly associated with macromolecular complexes and also to be resistant to solubilization with detergent, indicating that the kinase is not retained by membranes. The biochemical composition of the cytoplasmic complexes remains to be resolved. The fact that RNase treatment only leads to partial solubilization of HsPI3K-C2α from cytoplasmic complexes (Fig.2 C) argues that in the cytoplasm HsPI3K-C2α resides in two distinct pools. One pool appears to be associated with RNA, whereas the other could be indicative for HsPI3K-C2α bound to clathrin-coated vesicles (
      • Domin J.
      • Gaidarov I.
      • Smith M.E.
      • Keen J.H.
      • Waterfield M.D.
      ). It less probable that HsPI3K-C2α associates with the actin cytoskeleton or microtubules, since treatment of cells with the actin-depolymerizing compound cytochalasin D or microtubule-destabilizing conditions did not cause its solubilization.2
      Localization of HsPI3K-C2α to nuclear speckles implies some mechanism of nuclear import. A prerequisite for many proteins to be imported into the nucleus is a NLS. By expressing different segments of HsPI3K-C2α as GFP fusion proteins, we identified the highly basic sequence KRKTKISRKTR to be the determinant for the nuclear localization of HsPI3K-C2α. This sequence, which is located within the C2-like domain, contrasts the function of known C2 domains, which target proteins to the plasma membrane (
      • Nalefski E.A.
      • Falke J.J.
      ). The NLS of HsPI3K-C2α does not show strong similarity to other known nuclear-targeting sequences; rather, its overall composition of basic (7 of 11) amino acids provides some degree of homology (TableII). The best similarity was found with the NLS of rat ribosomal protein L31 (RLSRKR) (
      • Quaye I.K.
      • Toku S.
      • Tanaka T.
      ). The homology with the NLS of ribosomal protein may explain the pronounced nucleolar localization of HsPI3K-C2α-NLS-GFP and HsPI3K-C2α-NLS-DsRed fusion proteins (Fig. 5 C). The observation that the endogenous HsPI3K-C2α is not present in the nucleoli suggests that additional domains are required to target the protein to the nuclear speckles. Comparison of type II PI3-kinases from different species revealed that the sequence KRKTKISRKTR is conserved in HsPI3K-C2α and the murine PI3K-C2 kinases p170-m and cpk-m (Table I). This is in an agreement with our observation that in murine J7741.A and NIH 3T3 cell lines cpk-m localizes to nuclei. The human isoforms HsPI3K-C2β and HsPI3K-C2γ show partial homology in the region of NLS; however, only the tetrapeptide RKTK is fully conserved (Table I). It remains an open question of whether the corresponding sequences present in HsPI3K-C2β and HsPI3K-C2γ cause nuclear targeting. The fact that PI3K-C2β was found in nuclei of rat liver cells (
      • Sindic A.
      • Aleksandrova A.
      • Fields A.P.
      • Volinia S.
      • Banfic H.
      ) and the observation that rat PI3K-C2γ overexpressed in COS cells shows perinuclear localization (
      • Ono F.
      • Nakagawa T.
      • Saito S.
      • Owada Y.
      • Sakagami H.
      • Goto K.
      • Suzuki M.
      • Matsuno S.
      • Kondo H.
      ) suggests that translocation of different PI3K-C2 isoforms to the nucleus is mediated by the homologous sequences.
      Table IINuclear localization sequences
      Reference
      Minimal tetrapeptide consensusKBXB
      • Chelsky D.
      • Ralph R.
      • Jonak G.
      Bipartite domain of nucleoplasminKRPAATKKAGQAKKKKLD
      • Robbins J.
      • Dilworth S.M.
      • Laskey R.A.
      • Dingwall C.
      SV40 large T-antigenesPKKKRKV
      • Kalderon D.
      • Roberts B.L.
      • Richardson W.D.
      • Smith A.E.
      Ribosomal protein L31RLSRKR
      • Quaye I.K.
      • Toku S.
      • Tanaka T.
      HsPI3K-C2αKRKTKISRKTRThis study
      HsPI3K-C2βKRKTKVARKTCPotential NLS
      HsPI3K-C2γRRKTKSVPKCTPotential NLS
      MmPI3K-C2 (cpk-m)KRKTKISRKTRThis study
      MmPI3K-C2 (p170)KRKTKISRKTRThis study
      RexPKTRRRPRRSQRKRPPTP
      • Siomi H.
      • Shida H.
      • Nam S.H.
      • Nosaka T.
      • Maki M.
      • Hatanaka M.
      RevRQARRNRRRRWRERQR
      • Cochrane A.W.
      • Perkins A.
      • Rosen C.A.
      TatGRKKRRQRRRAHQ
      • Dang C.V.
      • Lee W.M.
      S6KRRRIALKKQRTKKNK
      • Heinze H.
      • Arnold H.H.
      • Fischer D.
      • Kruppa J.
      Mm PI3k-C2, Mus muluscus PI 3 kinase C2.
      The remarkable nuclear localization of HsPI3K-C2α suggests that phosphoinositides produced at speckles are critical regulators of nuclear activities. Two other PtdIns phosphate 5-kinases, the type I and II isoforms, which exhibit different substrate specificity, were also shown to associate with nuclear speckles (
      • Boronenkov I.V.
      • Loijens J.C.
      • Umeda M.
      • Anderson R.A.
      ). Upon inhibition of transcription, PtdIns phosphate 5-kinases redistribute identically with their product PtdIns (4,5)P2 and defined speckle components. The hypothesis that nuclear polyphosphoinositides are involved in regulation of transcription is further supported by the observation that the negatively charged lipids PtdIns (4,5)P2 and PtdIns 3,4,5-trisphosphate can reverse the inhibition of transcription caused by histone H1 (
      • Yu H.
      • Fukami K.
      • Watanabe Y.
      • Ozaki C.
      • Takenawa T.
      ) and that PtdIns (4,5)P2 is involved in BAF-mediated chromatin remodeling (
      • Zhao K.
      • Wang W.
      • Rando O.J.
      • Xue Y.
      • Swiderek K.
      • Kuo A.
      • Crabtree G.R.
      ). A specific role for PI (3,4)P2, the major product of HsPI3K-C2α, in nuclear events remains to be shown.

      Acknowledgments

      We thank our colleagues Dr. A. Rosen (Baltimore), Dr. R. Luhrmann (Goettingen, Federal Republic of Germany), Dr. M. Bachmann (Mainz, Federal Republic of Germany), and Dr. J. Domin (London, UK) for gifts of reagents, Dr. H. Langen (Basel, Switzerland) for mass spectrometry analysis, and Dr. G. Natoli and Dr. M. Molinari (Bellinzona, Switzerland) for critically reading the manuscript.

      REFERENCES

        • Fruman D.A.
        • Meyers R.E.
        • Cantley L.C.
        Annu. Rev. Biochem. 1998; 67: 481-507
        • Turner S.J.
        • Domin J.
        • Waterfield M.D.
        • Ward S.G.
        • Westwick J.
        J. Biol. Chem. 1998; 273: 25987-25995
        • Arcaro A.
        • Volinia S.
        • Zvelebil M.J.
        • Stein R.
        • Watton S.J.
        • Layton M.J.
        • Gout I.
        • Ahmadi K.
        • Downward J.
        • Waterfield M.D.
        J. Biol. Chem. 1998; 273: 33082-33090
        • Arcaro A.
        • Zvelebil M.J.
        • Wallasch C.
        • Ullrich A.
        • Waterfield M.D.
        • Domin J.
        Mol. Cell. Biol. 2000; 20: 3817-3830
        • Brown R.A.
        • Domin J.
        • Arcaro A.
        • Waterfield M.D.
        • Shepherd P.R.
        J. Biol. Chem. 1999; 274: 14529-14532
        • Sindic A.
        • Aleksandrova A.
        • Fields A.P.
        • Volinia S.
        • Banfic H.
        J. Biol. Chem. 2001; 276: 17754-17761
        • Zhang J.
        • Banfic H.
        • Straforini F.
        • Tosi L.
        • Volinia S.
        • Rittenhouse S.E.
        J. Biol. Chem. 1998; 273: 14081-14084
        • Newton A.C.
        Curr. Biol. 1995; 5: 973-976
        • Domin J.
        • Waterfield M.D.
        FEBS Lett. 1997; 410: 91-95
        • Wymann M.P.
        • Pirola L.
        Biochim. Biophys. Acta. 1998; 1436: 127-150
        • Misawa H.
        • Ohtsubo M.
        • Copeland N.G.
        • Gilbert D.J.
        • Jenkins N.A.
        • Yoshimura A.
        Biochem. Biophys. Res. Commun. 1998; 244: 531-539
        • Ono F.
        • Nakagawa T.
        • Saito S.
        • Owada Y.
        • Sakagami H.
        • Goto K.
        • Suzuki M.
        • Matsuno S.
        • Kondo H.
        J. Biol. Chem. 1998; 273: 7731-7736
        • Domin J.
        • Gaidarov I.
        • Smith M.E.
        • Keen J.H.
        • Waterfield M.D.
        J. Biol. Chem. 2000; 275: 11943-11950
        • Payrastre B.
        • Nievers M.
        • Boonstra J.
        • Breton M.
        • Verkleij A.J.
        • van Bergen en Henegouwen P.M.
        J. Biol. Chem. 1992; 267: 5078-5084
        • Boronenkov I.V.
        • Loijens J.C.
        • Umeda M.
        • Anderson R.A.
        Mol. Biol. Cell. 1998; 9: 3547-3560
        • Fu X.D.
        • Maniatis T.
        Nature. 1990; 343: 437-441
        • Spector D.L.
        • Fu X.D.
        • Maniatis T.
        EMBO J. 1991; 10: 3467-3481
        • Spector D.L.
        Curr. Opin. Cell Biol. 1993; 5: 442-447
        • Lewis J.D.
        • Tollervey D.
        Science. 2000; 288: 1385-1389
        • Phair R.D.
        • Misteli T.
        Nature. 2000; 404: 604-609
        • Sinclair G.D.
        • Brasch K.
        Exp. Cell Res. 1978; 111: 1-14
        • O'Keefe R.T.
        • Mayeda A.
        • Sadowski C.L.
        • Krainer A.R.
        • Spector D.L.
        J. Cell Biol. 1994; 124: 249-260
        • Spector D.L.
        • Schrier W.H.
        • Busch H.
        Biol. Cell. 1983; 49: 1-10
        • Misteli T.
        • Spector D.L.
        Mol. Biol. Cell. 1996; 7: 1559-1572
        • Gui J.F.
        • Lane W.S.
        • Fu X.D.
        Nature. 1994; 369: 678-682
        • Dingwall C.
        • Laskey R.A.
        Curr. Biol. 1998; 8: 922-924
        • Virbasius J.V.
        • Guilherme A.
        • Czech M.P.
        J. Biol. Chem. 1996; 271: 13304-13307
        • Domin J.
        • Pages F.
        • Volinia S.
        • Rittenhouse S.E.
        • Zvelebil M.J.
        • Stein R.C.
        • Waterfield M.D.
        Biochem. J. 1997; 326: 139-147
        • Didichenko S.A.
        • Tilton B.
        • Hemmings B.A.
        • Ballmer-Hofer K.
        • Thelen M.
        Curr. Biol. 1996; 6: 1271-1278
        • Thelen M.
        • Wymann M.P.
        • Langen H.
        Proc. Natl. Acad. Sci. U. S. A. 1994; 91: 4960-4964
        • Stephens L.E.
        • Eguinoa A.
        • Erdjument-Bromage H.
        • Lui M.
        • Cooke F.
        • Coadwell J.
        • Smrcka A.
        • Thelen M.
        • Cadwallader K.
        • Tempst P.
        • Hawkins P.T.
        Cell. 1997; 89: 105-114
        • Zini N.
        • Ognibene A.
        • Bavelloni A.
        • Santi S.
        • Sabatelli P.
        • Baldini N.
        • Scotlandi K.
        • Serra M.
        • Maraldi N.M.
        Histochem. Cell Biol. 1996; 106: 457-464
        • Lu P.J.
        • Hsu A.L.
        • Wang D.S.
        • Yan H.Y.
        • Yin H.L.
        • Chen C.S.
        Biochemistry. 1998; 37: 5738-5745
        • Metjian A.
        • Roll R.L.
        • Ma A.D.
        • Abrams C.S.
        J. Biol. Chem. 1999; 274: 27943-27947
        • Misteli T.
        • Spector D.L.
        Curr. Opin. Cell Biol. 1998; 10: 323-331
        • Fey E.G.
        • Krochmalnic G.
        • Penman S.
        J. Cell Biol. 1986; 102: 1654-1665
        • Bisotto S.
        • Lauriault P.
        • Duval M.
        • Vincent M.
        J. Cell Sci. 1995; 108: 1873-1882
        • Carmo-Fonseca M.
        • Pepperkok R.
        • Carvalho M.T.
        • Lamond A.I.
        J. Cell Biol. 1992; 117: 1-14
        • Colwill K.
        • Pawson T.
        • Andrews B.
        • Prasad J.
        • Manley J.L.
        • Bell J.C.
        • Duncan P.I.
        EMBO J. 1996; 15: 265-275
        • Gross S.D.
        • Loijens J.C.
        • Anderson R.A.
        J. Cell Sci. 1999; 112: 2647-2656
        • Misteli T.
        • Caceres J.F.
        • Spector D.L.
        Nature. 1997; 387: 523-527
        • Nalefski E.A.
        • Falke J.J.
        Protein Sci. 1996; 5: 2375-2390
        • Quaye I.K.
        • Toku S.
        • Tanaka T.
        Eur. J. Cell Biol. 1996; 69: 151-155
        • Yu H.
        • Fukami K.
        • Watanabe Y.
        • Ozaki C.
        • Takenawa T.
        Eur. J. Biochem. 1998; 251: 281-287
        • Zhao K.
        • Wang W.
        • Rando O.J.
        • Xue Y.
        • Swiderek K.
        • Kuo A.
        • Crabtree G.R.
        Cell. 1998; 95: 625-636
        • Chelsky D.
        • Ralph R.
        • Jonak G.
        Mol. Cell. Biol. 1989; 9: 2487-2492
        • Robbins J.
        • Dilworth S.M.
        • Laskey R.A.
        • Dingwall C.
        Cell. 1991; 64: 615-623
        • Kalderon D.
        • Roberts B.L.
        • Richardson W.D.
        • Smith A.E.
        Cell. 1984; 39: 499-509
        • Siomi H.
        • Shida H.
        • Nam S.H.
        • Nosaka T.
        • Maki M.
        • Hatanaka M.
        Cell. 1988; 55: 197-209
        • Cochrane A.W.
        • Perkins A.
        • Rosen C.A.
        J. Virol. 1990; 64: 881-885
        • Dang C.V.
        • Lee W.M.
        J. Biol. Chem. 1989; 264: 18019-18023
        • Heinze H.
        • Arnold H.H.
        • Fischer D.
        • Kruppa J.
        J. Biol. Chem. 1988; 263: 4139-4144