Bacterial lipopolysaccharide (LPS) produces rapid changes in macrophage protein synthesis and function. Phosphorylation of the 25 kDa mRNA cap-binding protein (eIF-4E) in model systems regulates the efficiency of protein synthesis. We report that both LPS and tumor necrosis factor-alpha (TNF-alpha) stimulate phosphorylation of eIF-4E and the p220 component of eIF-4F in bone marrow-derived macrophages. Moreover, anti-TNF-alpha antibodies inhibit LPS-stimulated phosphorylation of eIF-4E and p220 by 43% (+/- 6%) and 50% (+/- 5%), respectively. Our results indicate that LPS stimulates eIF-4F phosphorylation by a TNF-alpha-dependent mechanism, and suggest that phosphorylation of eIF-4F might play a role in the post-transcriptional regulation of gene expression in macrophages exposed to LPS.
A genetic selection procedure has been developed which makes the growth of E. coli dependent on expression of a cGMP phosphodiesterase cDNA. E. coli does not contain a cGMP phosphodiesterase, and guanine auxotrophs cannot extract the guanine from cGMP. If a functional cGMP phosphodiesterase is introduced, then guaA auxotrophs will grow on cGMP as a guanine source. The method also selects GMP synthetase cDNAs, which complement the guanine auxotrophy directly. Expression of a Dictyostelium discoideum or human heart cyclic nucleotide phosphodiesterase cDNA permits growth of the E. coli guaA auxotroph in the presence of cGMP. Several commercial cDNA libraries were corrupt and contained phosphodiesterase and/or GMP synthetase sequences that were from a contaminating DNA source.
Protein kinase C (PKC) is thought to play an important role in neuronal function by mediating changes in synaptic strength. Specifically, it has been argued that persistent PKC activation underlies the maintenance of long-term potentiation (LTP) of synaptic transmission in the hippocampus, a model widely used to study mammalian learning and memory. Because the induction of LTP is known to be dependent upon Ca2+ influx into the postsynaptic neuron, we investigated Ca(2+)-dependent mechanisms that operate to elicit persistent PKC activation in the hippocampus. Hippocampal homogenates were incubated with Ca2+ for a brief period and subsequently assayed for persistent changes in basal (Ca(2+)-independent) PKC activity, using the selective PKC substrate neurogranin(28-43) (NG(28-43)). After Ca2+ incubation, basal PKC phosphorylation of NG(28-43) was increased and expression of the increased activity could be inhibited by PKC(19-36), a selective peptide inhibitor of PKC. These data indicate the presence of a persistently activated form of PKC in Ca(2+)-pretreated hippocampal homogenates. The persistently activated PKC was localized to the soluble fraction of homogenates. Generation of the soluble, persistently activated form of PKC was blocked by the calpain inhibitor, leupeptin, suggesting a proteolytic activation of PKC. Column chromatography and Western blots indicated the presence of PKM, a proteolytic fragment of PKC that is active in the absence of calcium, diacylglycerols, or phospholipid cofactors. Thus, Ca2+ induces proteolytic activation of PKC in hippocampal homogenates. This suggests that proteolytic activation is a plausible candidate as a mechanism underlying the persistent activation of PKC associated with LTP.
The correlation between phospholamban and sarcoplasmic reticulum Ca(2+)-transporting ATPase levels and the magnitude of phospholamban-mediated stimulation of sarcoplasmic reticulum Ca2+ transport was examined in microsomes prepared from rabbit and canine cardiac, slow twitch and fast twitch skeletal muscle. Phospholamban was absent from microsomes prepared from fast twitch skeletal muscle but present at comparable levels in microsomes prepared from cardiac and slow twitch skeletal muscle. Levels of Ca(2+)-transporting ATPase were higher in microsomes prepared from slow twitch skeletal muscle than in microsomes prepared from cardiac muscle, however, and ratios of phospholamban to Ca(2+)-transporting ATPase were several fold greater in microsomes prepared from cardiac muscle than in microsomes prepared from slow twitch skeletal muscle. Stimulation of ATP-dependent Ca2+ transport following phosphorylation of phospholamban by cAMP-dependent protein kinase or incubation with anti-phospholamban monoclonal antibody was observed only in cardiac muscle microsomes. These observations indicate that phospholamban, while present in both cardiac and slow twitch skeletal muscle, may be involved in the hormonal regulation of sarcoplasmic reticulum Ca2+ transport only in the former, and that the lack of phospholamban-mediated stimulation of Ca2+ transport in slow twitch skeletal muscle sarcoplasmic reticulum may result from the lower ratio of phospholamban to Ca(2+)-transporting ATPase in this tissue.
Rat parathyroid hormone (PTH) stimulates cAMP-dependent protein kinase and protein kinase C activity in the kidney. However, PTH increases intracellular Calcium in primary cultures of proximal tubular cells. We have investigated the possibility that PTH also stimulates Calcium/calmodulin-dependent protein kinase II (CaM kinase II). We have employed the tandem chromatographic column method, using synthetic peptide as a substrate, to measure the renal CaM kinase II activity. PTH (250 nM) stimulated CaM kinase II activity by about 50% after 15 sec., and activity returned to baseline by 2 min. Calmodulin antagonists significantly impaired the stimulatory action of PTH whereas basal levels of CaM kinase II activity were relatively unaffected. This study demonstrates that PTH does activate CaM kinase II in renal tissue, and suggests another pathway for the actions of PTH in the kidney.
We have identified multiple members of the protein tyrosine phosphatase family in three subcellular compartments from rat liver; membrane, cytoskeleton and cytosol. Characterization based on substrate specificity, size, and reactivity with an anti-peptide antiserum against human placental PTP1B indicate the presence of at least three PTPases in Triton X-100 extracts of particulate membranes. Of these, one of 600 kDa possesses characteristics of a transmembrane, receptor-like enzyme. A fourth particulate PTPase (70 kDa) represents a distinct cytoskeletal PTPase. Cytosol contains one main PTPase species which was detected as a 41 kDa protein in Western immunoblots. These data indicate the existence of multiple hepatic PTPases whose differences in structure and subcellular localization may reflect functional heterogeneity.
Activation of G-proteins by internal perfusion with GTP-gamma-S or external application of carbachol resulted in oscillations of cytoplasmic Ca2+ in isolated mouse pancreatic beta-cells. The Ca2+ transients were associated with the generation of K+ currents sufficiently pronounced to induce marked pulses of hyperpolarization. The oscillatory G-protein response remained largely unaffected when altering the membrane potential. The oscillations became less frequent in the presence of 1 mM neomycin and disappeared when the cells were internally perfused with 100 micrograms/ml heparin. The frequency of the oscillations was positively correlated with the basal level of cytoplasmic Ca2+. Addition of Ca2+ to the internal perfusion medium increased the oscillatory rate and buffering of the ion with Indo-1 or EGTA had the opposite effect. It is concluded that G-protein activation results in cyclic mobilisation of intracellular calcium mediated by inositol-1,4,5-triphosphate and that the basal concentration of cytoplasmic Ca2+ is an important determinant for the frequency of the oscillations.
Several lines of evidence indicate that phosphorylation of the 25 kDa mRNA cap binding protein (eIF-4E) stimulates the efficiency of translational initiation. While the protein kinases which catalyze this reaction in intact cells have not been completely identified, evidence suggests that protein kinase C phosphorylates serine residues of eIF-4E in intact cells. In this study we demonstrate that protein kinase C also phosphorylates threonine residues of recombinant human eIF-4E in vitro. Phosphorylation of threonine and serine was observed over a range of eIF-4E and salt concentrations. However, relatively low levels of phosphorylation were seen even under optimal conditions. Similar results were observed with native eIF-4E purified from human erythrocytes. These findings demonstrate that protein kinase C can phosphorylate both serine and threonine residues of eIF-4E in vitro, but suggest that protein kinase C may not be the primary enzyme that phosphorylates eIF-4E in vivo.
We have examined the ability of recombinant human epidermal growth factor (EGF) and bradykinin (BK) to stimulate formation of inositol polyphosphates and sn-1,2-diacylglycerol (DAG), and mobilize intracellular Ca2+ ([Ca2+]i) in adult human keratinocytes (KC). Inositol polyphosphates were resolved by high performance liquid chromatography coupled with flow detector spectroscopy. Free intracellular calcium was quantitated using digital ratio imaging fluorescence microscopy of fura-2 loaded KC. The mass amount of DAG was quantitated using the DAG kinase reaction. When comparing maximal doses of BK (0.1-mu-M) and EGF (200 ng/ml), BK stimulated larger increases in all second messengers measured. The majority of cells responded rapidly to BK with global increases in [Ca2+]i. Cells responding to EGF were fewer in number and slower to respond with the Ca2+ signal being less pronounced. Treatment of cells with pertussis toxin (PTX) for 24 h significantly attenuated the BK-stimulated inositol polyphosphate formation and [Ca2+]i while the EGF response remained unaffected in both parameters. BK (10(-9) to 10(-6) M) did not stimulate DNA synthesis in KC as measured by [H-3]-thymidine incorporation when cultures were treated for 5 days. These results demonstrate that the-coupling and biochemical signals produced by stimulation of BK and EGF receptors in human KC are different and suggests that stimulation of second messenger formation from inositol lipid hydrolysis may not be an absolute requirement for the initiation of cell proliferation.
Mitotic spindles isolated from prometaphase-arrested mammalian cells contain associated protein kinases that are extracted by high salt treatment. Their fractionation by ion-exchange chromatography reveals three major peaks of protein kinase activity that phosphorylate brain microtubule-associated proteins and differ in their substrate specificity. One of them has been identified as a casein kinaseII-like enzyme. A mitotic spindle-associated 325 kDa protein related to brain MAP1B is a major substrate for this casein kinase II-like enzyme. Another mitotic spindle protein kinase has been tentatively identified as a proline-directed protein kinase.
In two separate experiments, control and PGF2 alpha-treated unilaterally ovariectomized ewes received vehicle or adenosine every four hours through an exteriorized catheter installed either to parafuse the ovarian vascular pedicle of the remaining ovary containing a corpus luteum (CL) or to infuse intrauterine ipsilateral to the remaining ovary with the CL. Infusions were given every four hours from day 7 through day 22 postestrus or until ewes returned to estrus. Chronic infusion of vehicle intrauterine or parafusion of the ovarian vascular pedicle with vehicle in PGF2 alpha-treated ewes shortened the interestrous interval compared to controls (P less than 0.05). However, chronic parafusion of the ovarian vascular pedicle with adenosine in PGF2 alpha-treated ewes prevented a PGF2 alpha-induced premature luteolysis (P less than 0.05) but not when adenosine was infused chronically intrauterine (P greater than 0.05). It is concluded that adenosine can block a PGF2 alpha-induced premature luteolysis in vivo and may have roles in regulation of luteal secretion of progesterone but it is probably not the embryonic antiluteolysin of early pregnancy in ewes.
The effect of platelet pretreatment with moderate amounts of 1,2-dioctanoylglycerol on subsequent thrombin-induced activation and its correlation with the degree of protein phosphorylation is studied. Protein kinase C preactivation (treatment with 1 microM dioctanoylglycerol for 20 min) significantly reduces thrombin-promoted platelet aggregation, cytosolic calcium-rise, ATP-secretion and, albeit to a lesser extent, protein phosphorylation. Exposure of platelets to dioctanoylglycerol brings about a transient phosphorylation of a 47 kDa protein and a slight but more persistent phosphorylation of proteins of approximate molecular mass 26 and 68 kDa. It is hypothesized that the latter phosphoproteins are responsible for the inhibition of the thrombin-promoted platelet activation. Agonist-evoked aggregation is more affected by a long (20 min) rather than a short (1 min) pretreatment with dioctanoylglycerol, showing no correlation with the phosphorylation of the major substrates of protein kinase C.
We have shown that growth of S49 WT mouse lymphoma cells for 24 hr in 3 nM epinephrine produced very significant desensitization of subsequent cellular cAMP responses to challenges with higher concentrations of epinephrine. The effects of this long-term treatment (LTT) were obvious in intact cells and also when adenylate cyclase activity was measured in semi-purified membranes. Beta 2-adrenergic receptors (beta 2AR) were decreased by LTT, and the desensitization of adenylate cyclase was due at least in part to this down-regulation. When mouse L cells transfected with WT beta 2AR from hamster lung (L-WT beta 2AR cells) were subjected to LTT, the attenuation of adenylate cyclase in membranes was obvious, but the consequences of LTT on intact L-WT beta 2AR cells were highly equivocal. That is, when the effects of epinephrine on cellular cAMP levels were measured in LTT or control L-WT beta 2AR cells little desensitization was apparent. Further, cellular cAMP excursions in response to even very high concentrations of epinephrine were very small in control L-WT beta 2AR cells as compared to control S49 WT cells. Subsequent experiments have shown that L-WT beta 2AR cells possess a phosphodiesterase (PDE) which demonstrates marked positive cooperativity with cAMP with a Hill coefficient of 2. The EC50 for cAMP hydrolysis was approximately 30 nM in cell free preparations. cGMP was a positive allosteric effector of the L-WT beta 2AR cell PDE. Further, when cellular cAMP levels in intact L-WT beta 2AR cells were raised above a threshold by treatment with 0.5 microM forskolin and 2 mM IBMX with the epinephrine challenge, the effect of LLT became obvious in the intact cell system. These data demonstrate that cAMP responses to hormones are greatly decreased in systems where the predominant PDE demonstrates positive cooperativity for cAMP.
A method has been developed for measuring specific protein phosphatase activity in biological samples using synthetic, phospho-Kemptide and phospho-GS-peptide. This method uses ion-exchange chromatography to determine phosphatase activity by quantifying the release of [32P]phosphate directly. The method was used to measure phosphatase activity of rat kidney, adrenals, heart, and liver cytosol and the activity of purified alkaline phosphatases, protein phosphatase 1, and protein phosphatase 2A. Ion-exchange chromatography was also used for the preparation of the radiolabeled phosphopeptide substrates. This method results in high recovery and specific activity of the labeled peptides. These techniques should be useful in isolating and characterizing specific protein phosphatases found in cells.
After photolabelling of purified bovine brain particulate cGMP stimulated phosphodiesterase (PDE) with [32P]cGMP, incubation with V8 Endoproteinase Glu-C produced several fragments; most of the [32P] was associated with smaller fragments (approximately 12-14 kDa), and some occasionally with larger fragments (approximately 55-57 kDa). Partial amino acid sequences were determined for the smaller photolabelled fragments and other peptides. On Western immunoblots, affinity-purified antibodies against a synthetic peptide with a sequence matching part of that of the approximately 12-14 kDa photolabelled material reacted with intact PDE and the approximately 12-14 kDa and approximately 55-57 kDa fragments. Several partial cDNA clones encoding the cGMP-stimulated PDE were isolated from a Lambda Zap II bovine brain cDNA library. Deduced amino acid sequence from one cDNA clone, lambda cGS 3-1, as well as the partial sequence of the approximately 12-14 kDa and other fragments, exhibited considerable identity with amino acid residues 311-921 of the cardiac cGMP-stimulated PDE (Trong et al., 1990), including the putative cGMP-binding domain (Charbonneau et al., 1990). These results further define this cGMP-binding domain and suggest that different cGMP-stimulated PDEs will exhibit considerable homology, at least in their cGMP-binding region(s) and catalytic domains.
Subcellular liver fractions from rats receiving a subcutaneous injection of turpentine, which causes a local inflammation, show an increased synthesis of Prostaglandin E2 and Prostaglandin F2 alpha which reaches a peak 90 minutes and 3 hours after treatment, respectively. Stimulation of phospholipase A2 activity of liver cell preparations seems to be responsible for the supply of arachidonic acid necessary to feed PG synthesis: this stimulation is accompanied by unchanged levels of diacylglycerol lipase, diacylglycerol kinase and protein kinase C activities and by an unchanged content of diacylglycerol in the liver tissue. This picture does not favour the hypothesis of an involvement of phospholipase C in the early stages after turpentine treatment. Determinations of GTP-ase activity in plasma membrane-rich liver preparations give ambiguous results, which do not allow any conclusion on the possible role of G-proteins in phospholipase A2 activation.
Cyclic GMP mediates vascular smooth muscle relaxation to a variety of drugs and naturally-occurring substances. The reduction of intracellular Ca2+ levels is believed to underlie this action, but the mechanism of this effect is unknown. In order to test the hypothesis that inhibition of guanine nucleotide-binding protein function is involved in the actions of cGMP, the effects of cGMP-dependent protein kinase on the phosphorylation of both pertussis toxin-sensitive (G(i)/G(o)) and insensitive (G(z)) G-proteins were examined in vitro. None of these proteins were effective substrates for either cGMP- or cAMP-dependent protein kinases, despite the fact that assay conditions were designed to detect poorly phosphorylated substrate proteins. In line with these observations, atriopeptin II did not inhibit angiotensin II-treated inositol phosphate formation in cultured vascular smooth muscle cells. These results suggest that phosphorylation by cGMP-dependent protein kinase of these G-proteins is not the major mechanism by which cGMP reduces intracellular Ca2+ levels in vascular smooth muscle.
Tyrosine kinase activity was studied in the crude cytosolic and particulate fraction of normal mucosa and squamous cell carcinomas of the upper aero-digestive tract. In the presence of exogenously added phosphorylation substrate (Glu,Tyr4:1), the cytosolic tyrosine kinase activity was 6-fold higher in tumors compared to normal mucosa (p = 0.001), and in the particulate fraction the increase was 8-fold in tumors compared to normal mucosa. Different proposed tyrosine kinase inhibitors, including genistein, quercetin and the alpha-cyanocinnamide ST 638, were tested for their ability to inhibit phosphorylation of the synthetic tyrosine phosphorylation substrate. Phosphorylation of Glu,Tyr4:1 in tumors (cytosolic fraction) was reduced to 77.8 +/- 8.7% of the control value by 10 microM ST 638 (p less than 0.05), and to 50.7 +/- 10.4% by 100 microM quercetin (p less than 0.01). In normal mucosa (cytosolic fraction) the corresponding values were 41.7 +/- 16.6% in the presence of 10 microM ST 638 (p less than 0.05) and 32.1 +/- 5.8% in the presence of 100 microM quercetin (p less than 0.05). These inhibitors had no effect on the tyrosine kinase activity in the particulate fractions. Phosphorylation of endogenous proteins in the crude cytosolic fraction was evaluated by SDS-polyacrylamide gel electrophoresis after alkali treatment of the gels. Autoradiography of the gels treated in this manner revealed bands corresponding to phosphorylated proteins with apparent molecular weight of 18, 23, 37-38, 42-44 (double band), 53-55 (double band), 61 and 92-94 (double band) kD. Quercetin (100 microM) markedly reduced the phosphorylation of these proteins, while no effect of ST 638 could be seen. Heparin (20 micrograms/ml) stimulated the phosphorylation of three proteins with apparent molecular weight of 39 and about 72 kD, respectively, and inhibited the phosphorylation of 2 proteins with molecular weight of 92 and 53 kD in tumors. These features were observed in both tumors and normal tissue, with the exception that heparin only stimulated the 72 kD band in normal mucosa and that the phosphorylation was markedly higher in tumors. In summary, our results show an increased tyrosine phosphorylation in squamous cell carcinomas of the upper aero-digestive tract compared to normal oral mucosa. These differences and their origin might be of vital importance in the regulation of events leading to malignant transformation.
Since inositol 1,4,5-trisphosphate (1,4,5-IP3) and inositol 1,3,4,5-tetrakisphosphate (1,3,4,5-IP4) have been described to modulate Ca(2+)-channels, we investigated the possible participation of 1,4,5-IP3 and/or 1,3,4,5-IP4 in the bradykinin-induced Ca(2+)-influx into cultured porcine aortic endothelial cells. In our experiments bradykinin induced a quick release of Ca2+ from intracellular stores and a longlasting Ca(2+)-influx, which remained constant for at least 15 minutes. In contrast to its effect on [Ca2+]i, bradykinin only transiently elevated 1,4,5-IP3 and 1,3,4,5-IP4 levels. Ten minutes after addition of bradykinin, both 1,4,5-IP3 and 1,3,4,5-IP4 levels returned to basal values, whereas Ca(2+)-influx was still unaltered. Furthermore, preincubation of endothelial cell with phorbol-12-myristate-13-acetate (PMA) abolished the stimulatory effect of bradykinin on the formation of 1,4,5-IP3 and 1,3,4,5-IP4, but did not affect the longlasting Ca(2+)-influx. These data provide evidence that in endothelial cells inositolphosphates are not involved in the regulation of bradykinin-induced longlasting Ca(2+)-influx.
In streptolysin O permeabilized acini that were normally responsive to carbamylcholine and cholecystokinin octapeptide, amylase secretion was stimulated: a) by the stable guanyl nucleotides with a potency decreasing as follows: guanosine 5'-O-(3-thiotriphosphate) (GTP gamma S) greater than guanylyl-imidodiphosphate (GMP-PNP) = guanylyl-(beta, gamma-methylene)-diphosphonate (GMP-PCP), in the presence of 0.5 mM calcium and b) by calcium alone (EC50 3 microM). The maximal secretory effect of calcium alone (a 2-fold increase) was less effective than that of GTP gamma S and calcium offered in combination (an 8-fold increase). In the virtual absence of Ca2+, GTP gamma S still stimulated amylase release (a 3-fold increase) while 12-O-tetradecanoylphorbol 13-acetate (TPA) did not. The relative potencies of guanyl nucleotides were GTP gamma S greater than GMP-PNP = GMP-PCP = GTP on phosphatidylinositol 4,5-bisphosphate (PIP2) breakdown, GTP gamma S greater than GMP-PNP greater than GMP-PCP = GTP on 45Ca2+ efflux, and GTP GMP-PNP = GMP-PCP = GTP gamma S on [1-14C]arachidonate efflux. Based on these data, the contribution of G proteins to stimulus-secretion coupling beyond the transduction of receptor signal is considered.
