Inositides in the nucleus: taking stock of PLC beta 1

Phospholipase C beta 4 in mouse hepatocytes: rhythmic expression and cellular distribution

BACKGROUND

Circadian regulated physiological processes have been well documented in the mammalian liver. Phospholipases are important mediators of both cytoplasmic and nuclear signaling mechanisms in hepatocytes, and despite a potentially critical role for these enzymes in regulating the temporal aspect of hepatic physiology, their involvement in the circadian liver clock has not been the subject of much investigation. The phospholipase C beta4 (PLCbeta4) enzyme is of particular interest as it has been linked to circadian clock function. In general, there is no knowledge of the role of the PLCbeta4 isozyme in mammalian hepatocytes as this is the first report of its expression in the mammalian liver.

RESULTS

We found that in the liver of mice housed on a light:dark cycle, PLCbeta4 protein underwent a significant circadian rhythm with a peak occurring during the early night. In constant darkness, the protein rhythm was more robust and peaked around dusk. We also observed a significant oscillation in plcbeta4 gene expression in the livers of mice housed in both photoperiodic and constant dark conditions. The cellular distribution of the protein in hepatocytes varied over the course of the circadian day with PLCbeta4 primarily cytoplasmic around dusk and nuclear at dawn.

CONCLUSION

Our results indicate that PLCbeta4 gene and protein expression is regulated by a circadian clock in the mouse liver and is not dependent on the external photoperiod. A light-independent daily translocation of PLCbeta4 implies that it may play a key role in nuclear signaling in hepatocytes and serve as a daily temporal cue for physiological processes in the liver.

Regulation of nuclear processes by inositol polyphosphates

Inositide signaling pathways represent a multifaceted ensemble of cellular switches capable of regulating a number of processes, for example, intracellular calcium release, membrane trafficking, chemotaxis, ion channel activity and several nuclear functions. Over 30 inositide messengers are found in eukaryotic cells that may be grouped into two classes: (1) inositol lipids, phosphatidylinositols or phosphoinositides (PIPs) and (2) water-soluble inositol polyphosphates (IPs). This review will focus on inositol polyphosphate kinases (IPK) and inositol pyrophosphate synthases (IPS) responsible for the cellular production of IP(4), IP(5) IP(6) and PP-IPs. Of interest, IPK and IPS proteins localize, in part, within the nucleus and their activities are necessary for proper regulation of gene expression, mRNA export, DNA repair and telomere maintenance. The breadth of nuclear processes regulated and the evolutionary conservation of the genes involved in their synthesis have sparked renewed interest in inositide messengers derived from sequential phosphorylation of inositol 1,4,5-trisphosphate.

Mesangial cell-reduced Ca2+signaling in high glucose is due to inactivation of phospholipase C-β3by protein kinase C

In high glucose, glomerular mesangial cells (MCs) demonstrate impaired Ca2+signaling in response to seven-transmembrane receptor stimulation. To identify the mechanism, we first postulated decreased release from intracellular stores. Intracellular Ca2+was measured in fluo-3-loaded primary cultured rat MCs using confocal fluorescence microscopy. In high glucose (HG) 30 mM for 48 h, the 25 nM ionomycin-stimulated intracellular Ca2+response was reduced to 82% of that observed in normal glucose (NG). In NG 5.6 mM, Ca2+responses to endothelin (ET)-1 and platelet-derived growth factor (PDGF) were unchanged in cells cultured in 50 nM Ca2+vs. 1.8 mM Ca2+. Depletion of intracellular Ca2+stores with thapsigargin eliminated ET-1-stimulated Ca2+responses. Incubation in 30 mM glucose (HG) for 48 h or stimulation with phorbol myristate acetate (PMA) for 10 min eliminated the Ca2+response to ET-1 but had no effect on the PDGF response. Downregulation of protein kinase C (PKC) with 24-h PMA or inhibition with Gö6976 in HG normalized the Ca2+response to ET-1. Because ET-1 and PDGF stimulate Ca2+signaling through different phospholipase C pathways, we hypothesized that, in HG, PKC selectively phosphorylates and inhibits PLC-β3. Using confocal immunofluorescence imaging, in NG, a 1.6- to 1.7-fold increase in PLC-β3Ser1105phosphorylation was observed following PMA or ET-1 stimulation for 10 min. In HG, immunofluorescent imaging and immunoblotting showed increased PLC-β3phosphorylation, without change in total PLC-β3, which was reversed with 24-h PMA or Gö6976. We conclude that reduced Ca2+signaling in HG cannot be explained by reduced Ca2+stores but is due to conventional PKC-dependent phosphorylation and inactivation of PLC-β3.

The structure of phosphatidylinositol transfer protein alpha reveals sites for phospholipid binding and membrane association with major implications for its function

Elucidation of the three-dimensional structure of phosphatidylinositol transfer protein alpha (PI-TPalpha) void of phospholipid revealed a site of membrane association connected to a channel for phospholipid binding. Near the top of the channel specific binding sites for the phosphorylcholine and phosphorylinositol head groups were identified. The structure of this open form suggests a mechanism by which PI-TPalpha preferentially binds PI from a membrane interface. Modeling predicts that upon association of PI-TPalpha with the membrane the inositol moiety of bound PI is accessible from the medium. Upon release from the membrane PI-TPalpha adopts a closed structure with the phospholipid bound fully encapsulated. This structure provides new insights as to how PI-TPalpha may play a role in PI metabolism.

Fear conditioning-induced alterations of phospholipase C-beta1a protein level and enzyme activity in rat hippocampal formation and medial frontal cortex

We investigated the effects of one-trial fear conditioning on phospholipase C-beta1a catalytic activity and protein level in hippocampal formation and medial frontal cortex of untreated control rats and rats prenatally exposed to ethanol. One hour following fear conditioning of untreated control rats, phospholipase C-beta1a protein level was increased in the hippocampal cytosolic fraction and decreased in the hippocampal membrane and cortical cytosolic and cortical membrane fractions. Twenty-four hours after fear conditioning, phospholipase C-beta1a protein level was reduced in the hippocampal cytosolic fraction and elevated in the cortical nuclear fraction; in addition, 24 h after conditioning, phospholipase C-beta1a activity in the cortical cytosolic fraction was increased. Rats that were exposed prenatally to ethanol displayed attenuated contextual fear conditioning, whereas conditioning to the acoustic-conditioned stimulus was not different from controls. In behavioral control (unconditioned) rats, fetal ethanol exposure was associated with reduced phospholipase C-beta1a enzyme activity in the hippocampal nuclear, cortical cytosolic, and cortical membrane fractions and increased phospholipase C-beta1a protein level in the hippocampal membrane and cortical cytosolic fractions. In certain cases, prenatal ethanol exposure modified the relationship between fear conditioning and changes in phospholipase C-beta1a protein level and/or activity. The majority of these effects occurred 1 h, rather than 24 h, after fear conditioning. Multivariate analysis of variance revealed interactions between fear conditioning, subcellular fraction, and prenatal ethanol exposure for measures of phospholipase C-beta1a protein level in hippocampal formation and phospholipase C-beta1a enzyme activity in medial frontal cortex. In the majority of cases, fear conditioning-induced changes in hippocampal phospholipase C-beta1a protein level were augmented in rats prenatally exposed to ethanol. In contrast, fear conditioning-induced changes in cortical phospholipase C-beta1a activity were, often, in opposite directions in prenatal ethanol-exposed compared to diet control rats. We speculate that alterations in subcellular phospholipase C-beta1a catalytic activity and protein level contribute to contextual fear conditioning and that learning deficits observed in rats exposed prenatally to ethanol result, in part, from dysfunctions in phospholipase C-beta1a signal transduction.

Protein kinase C alpha -mediated negative feedback regulation is responsible for the termination of insulin-like growth factor I-induced activation of nuclear phospholipase C beta1 in Swiss 3T3 cells

Previous studies from several independent laboratories have demonstrated the existence of an autonomous phosphoinositide (PI) cycle within the nucleus, where it is involved in both cell proliferation and differentiation. Stimulation of Swiss 3T3 cells with insulin-like growth factor-I (IGF-I) has been shown to induce a transient and rapid increase in the activity of nuclear-localized phospholipase C (PLC) beta1, which in turn leads to the production of inositol trisphosphate and diacylglycerol in the nucleus. Nuclear diacylglycerol provides the driving force for the nuclear translocation of protein kinase C (PKC) alpha. Here, we report that treatment of Swiss 3T3 cells with Go6976, a selective inhibitor of PKC alpha, caused a sustained elevation of IGF-I-stimulated nuclear PLC activity. A time course study revealed an inverse relationship between nuclear PKC activity and the activity of nuclear PLC in IGF-I-treated cells. A time-dependent association between PKC alpha and PLC beta1 in the nucleus was also observed following IGF-I treatment. Two-dimensional phosphopeptide mapping and site-directed mutagenesis demonstrated that PKC promoted phosphorylation of PLC beta1 at serine 887 in the nucleus of IGF-I-treated cells. Overexpression of either a PLC beta1 mutant in which the PKC phosphorylation site Ser(887) was replaced by alanine, or a dominant-negative PKC alpha, resulted in a sustained activation of nuclear PLC following IGF-I stimulation. These results indicate that a negative feedback regulation of PLC beta1 by PKC alpha plays a critical role in the termination of the IGF-I-dependent signal that activates the nuclear PI cycle.

Regulation of phosphoinositide-specific phospholipase C

Eleven distinct isoforms of phosphoinositide-specific phospholipase C (PLC), which are grouped into four subfamilies (beta, gamma, delta, and epsilon), have been identified in mammals. These isozymes catalyze the hydrolysis of phosphatidylinositol 4,5-bisphosphate [PtdIns(4,5)P2] to inositol 1,4,5-trisphosphate and diacylglycerol in response to the activation of more than 100 different cell surface receptors. All PLC isoforms contain X and Y domains, which form the catalytic core, as well as various combinations of regulatory domains that are common to many other signaling proteins. These regulatory domains serve to target PLC isozymes to the vicinity of their substrate or activators through protein-protein or protein-lipid interactions. These domains (with their binding partners in parentheses or brackets) include the pleckstrin homology (PH) domain [PtdIns(3)P, beta gamma subunits of G proteins] and the COOH-terminal region including the C2 domain (GTP-bound alpha subunit of Gq) of PLC-beta; the PH domain [PtdIns(3,4,5)P3] and Src homology 2 domain [tyrosine-phosphorylated proteins, PtdIns(3,4,5)P3] of PLC-gamma; the PH domain [PtdIns(4,5)P2] and C2 domain (Ca2+) of PLC-delta; and the Ras binding domain (GTP-bound Ras) of PLC-epsilon. The presence of distinct regulatory domains in PLC isoforms renders them susceptible to different modes of activation. Given that the partners that interact with these regulatory domains of PLC isozymes are generated or eliminated in specific regions of the cell in response to changes in receptor status, the activation and deactivation of each PLC isoform are likely highly regulated processes.

A role for nuclear phospholipase Cbeta 1 in cell cycle control

Phosphoinositide signaling resides in the nucleus, and among the enzymes of the cycle, phospholipase C (PLC) appears as the key element both in Saccharomyces cerevisiae and in mammalian cells. The yeast PLC pathway produces multiple inositol polyphosphates that modulate distinct nuclear processes. The mammalian PLCbeta(1), which localizes in the nucleus, is activated in insulin-like growth factor 1-mediated mitogenesis and undergoes down-regulation during murine erythroleukemia differentiation. PLCbeta(1) exists as two polypeptides of 150 and 140 kDa generated from a single gene by alternative RNA splicing, both of them containing in the COOH-terminal tail a cluster of lysine residues responsible for nuclear localization. These clues prompted us to try to establish the critical nuclear target(s) of PLCbeta(1) subtypes in the control of cell cycle progression. The results reveal that the two subtypes of PLCbeta(1) that localize in the nucleus induce cell cycle progression in Friend erythroleukemia cells. In fact when they are overexpressed in the nucleus, cyclin D3, along with its kinase (cdk4) but not cyclin E is overexpressed even though cells are serum-starved. As a consequence of this enforced expression, retinoblastoma protein is phosphorylated and E2F-1 transcription factor is activated as well. On the whole the results reveal a direct effect of nuclear PLCbeta(1) signaling in G(1) progression by means of a specific target, i.e. cyclin D3/cdk4.

Inositides in the nucleus: further developments on phospholipase C beta 1 signalling during erythroid differentiation and IGF-I induced mitogenesis

Inositol lipids originally shown to be metabolized in the cytosol have been detected also in the nucleus, where they are both synthesized and hydrolyzed. In the case of erythroid differentiation of murine erythroleukemia cells (Friend cells) it has been previously shown that PLC beta 1, which is the major nuclear PLC, undergoes down-regulation upon treatment with DMSO or tiazofurin which act as differentiative agents. On the contrary, i.e., during IGF-I induced mitogenesis, it has been shown that PLC beta 1 is rapidly activated and this event is essential for the onset of DNA synthesis. Even though its key role in cell growth has been shown, both the mechanism by which nuclear PLC beta 1 is activated and the direct relationship with erythroid differentiation are still unknown. We have addressed the question if PLC beta 1 expression and activity in the nucleus are directly related or not to the establishment of the differentiated state and we have checked the two main ways of activation, i.e., via G-protein or via phosphorylation, in order to establish whether nuclear PLC beta 1 is regulated the same way as the one at the plasma membrane or not. The data reported here show that nuclear PLC beta 1 is responsible for a continuous recycling of Friend cells, acting as a negative regulator of differentiation and that its activation is dependent on the phosphorylation state.

Inositides in the nucleus: presence and characterisation of the isozymes of phospholipase beta family in NIH 3T3 cells

Previous reports from our laboratories and others have hinted that the nucleus is a site for an autonomous signalling system acting through the activation of the inositol lipid cycle. Among phospholipases (PLC) it has been shown previously that PLCbeta1 is specifically localised in the nucleus as well as at the plasma membrane. Using NIH 3T3 cells, it has been possible to obtain, with two purification strategies, in the presence or in the absence of Nonidet P-40, both intact nuclei still maintaining the outer membrane and nuclei completely stripped of their envelope. In these nuclei, we show that not only PLCbeta1 is present, but also PLCbeta2, PLCbeta3 and PLCbeta4. The more abounding isoform is PLCbeta1 followed by PLCbeta3, PLCbeta2 and PLCbeta4, respectively. All the isoforms are enriched in nuclear preparations free from nuclear envelope and cytoplasmatic debris, indicating that the actual localisation of the PLCbeta isozymes is in the inner nuclear compartment.