BBA - Molecular Cell Research (v.1497, #3)

Interleukin-1β increases binding of the iron regulatory protein and the synthesis of ferritin by increasing the labile iron pool by Domingo J. Piñero; Jing Hu; Braden M. Cook; Russell C. Scaduto; James R. Connor (279-288).
This study was undertaken to begin to elucidate the mechanisms by which cytokines influence intracellular iron homeostasis. Intracellular iron homeostasis is maintained by the coordinated regulation of ferritin and transferrin receptor synthesis. The synthesis of these proteins is coordinated by cytoplasmic iron regulatory proteins (IRP), which bind to iron responsive elements (IRE) on their mRNAs. We evaluated the effects of interleukin-1β (IL-1β) on iron metabolism in human astrocytoma cells (SW1088). Exposure to IL-1β for 16 h increased binding of the IRPs to the IRE and also increased ferritin synthesis. Using the iron sensitive dye calcein, we determined that the intracellular labile iron pool increased within 4 h of IL-1β exposure and continued to increase for 8 h, returning to normal by 16 h. We propose that the cytokine induced increase in the labile iron pool stimulates ferritin synthesis resulting in a subsequent decrease in the labile iron pool. The decrease in the labile iron pool is consistent with the increase in IRE/IRP interaction measured at 16 h. These results indicate that cytokines can influence the labile iron pool and the post-transcriptional regulatory mechanism for maintaining iron homeostasis. These results contribute to understanding the response of ferritin to inflammation by suggesting ferritin synthesis may reflect changes in the labile iron pool. The approach used in this study may provide a model system for studying relations between the labile iron pool and proteins responsible for maintaining intracellular homeostasis
Keywords: Iron regulation; Post-transcriptional regulation; mRNA; Cytokine; Fluorometry;

Biphasic effect of protein kinase C activators on spontaneous and glucocorticoid-induced apoptosis in primary mouse thymocytes by Jörg Rennecke; K.Hartmut Richter; Sabine Häussermann; Luise Stempka; Susanne Strand; Michael Stöhr; Friedrich Marks (289-296).
Spontaneous and glucocorticoid (fluocinolone acetonide, FA)-induced apoptosis of primary mouse thymocytes was inhibited by protein kinase C (PKC) activators such as bryostatin-1 and phorbol ester 12-O-tetradecanoyl-phorbol-13 acetate (TPA) within the first 2–4 h of incubation but was enhanced upon prolonged treatment. Only the anti-apoptotic but not the pro-apoptotic effect of TPA was completely suppressed by the PKC inhibitor Goe 6983 and moderately inhibited by Goe 6976. Immunoblot analysis revealed distinct PKC α, β, δ, η, θ, μ and ζ signals, a very faint PKCϵ and no PKCγ signal. Upon prolonged TPA treatment all PKC isoenzymes became downregulated, albeit at different rates (PKCδ>α>μ>β,θ≫η,ζ). No significant generation of caspase-derived catalytic PKC fragments, as found to be produced upon induction of apoptosis and to be pro-apoptotic in other systems, was observed in FA- or TPA-treated thymocytes. It is concluded that the early anti-apoptotic effect of TPA depends on the activation of n-type PKC isoenzymes, whereas stimulation of spontaneous and FA-induced apoptosis by TPA ensues, at least partially, from a downregulation (or inactivation) of anti-apoptotic PKC species, i.e. in primary thymocytes PKC activation is primarily involved in a negative regulation of apoptosis.
Keywords: Phorbol ester; Bryostatin-1; Protein kinase C; Apoptosis; Thymocyte; Glucocorticoid;

β2-Adaptin is constitutively de-phosphorylated by serine/threonine protein phosphatase PP2A and phosphorylated by a staurosporine-sensitive kinase by Jens Peter H Lauritsen; Charlotte Menné; Jesper Kastrup; Jes Dietrich; Niels Ødum; Carsten Geisler (297-307).
Clathrin-mediated endocytosis includes cycles of assembly and disassembly of the clathrin-coated vesicle constituents. How these cycles are regulated is still not fully known but previous studies have indicated that phosphorylation of coat subunits may play a role. Here we describe that β2-adaptin undergoes cycles of phosphorylation/de-phosphorylation in intact cells. Thus, β2-adaptin was constitutively de-phosphorylated by serine/threonine protein phosphatase 2A and phosphorylated by a staurosporine-sensitive kinase in vivo. Confocal laser scanning microscopy demonstrated that phosphorylated AP2 complexes were found more evenly distributed at the plasma membrane compared to non-phosphorylated AP2 complexes which were found in aggregates. Finally, we found that phosphorylation of β2-adaptin correlated with inhibition of clathrin-mediated endocytosis. Our results support the hypothesis that phosphorylation/de-phosphorylation of coat proteins plays a regulatory role in the assembly/disassembly cycle of clathrin-coated vesicles.
Keywords: Adaptin; Adaptor; Phosphorylation; Phosphatase; Kinase; Endocytosis;

Fluorescence imaging of pyrene-labeled lipids in living cells by Kimmo Tanhuanpää; Jorma Virtanen; Pentti Somerharju (308-320).
Microscopic imaging of fluorescent lipid derivatives is a powerful tool to study membrane organization and lipid trafficking but it is complicated by cellular autofluorescence background and photobleaching of the fluorophore as well as by the difficulty to selectively image membranes stacked on top of each other. Here we describe protocols that strongly alleviate such problems when pyrene-labeled lipids are being used. First, photobleaching of these lipids is virtually eliminated when oxygen is depleted from the medium by using a gentle and simple enzymatic method. Second, an image practically free of cellular autofluorescence contribution can be obtained simply by subtracting from the pyrene image the background image obtained at a slightly different excitation wavelength. This type of background subtraction more properly accounts for the typically uneven distribution of cellular background fluorescence than other, commonly used methods. Third, it is possible to selectively image the pyrene lipids in the plasma membrane by using plasma membrane-specific quencher trinitrophenyl lysophosphatidylethanolamine and image subtraction. Importantly, either the outer or the inner leaflet can be selectively imaged by labeling the cells with pyrene phosphatidylcholine or phosphatidylserine, respectively. These protocols should be of considerable help when studying organization of the plasma membrane or intracellular lipid trafficking.
Keywords: Fluorescence; Microscopy; Domain; Lipid traffic;

The release of amino acids from their vacuolar store was studied in situ, i.e. in cells with selectively permeabilized plasma membrane and functionally intact vacuoles. As we previously described [Roos et al., J. Biol. Chem. 272 (1997) 15849–15855], this transport process is regulated by extravacuolar adenylates at their physiological concentrations. We now show, using our test object Penicillium cyclopium, that not only purine but also pyrimidine nucleotides are involved in the control of efflux of vacuolar phenylalanine. At 0.1 mM adenosine or guanosine phosphates inhibit, whereas cytidine or uridine phosphates stimulate the rate of efflux. At 1 mM the same nucleotides have no measurable impact on efflux but abolish the effects of other nucleotides present at 0.1 mM. This argues for at least two interacting binding sites with different nucleotide affinities. The minimum structural requirement for any of the observed effects is a non-cyclic ribonucleoside monophosphate. In intact cells, cytosolic concentrations of ATP (representing purine nucleotides) and CTP (representing pyrimidine nucleotides) are 1–2 mM and 0.05–0.2 mM, respectively. ATP is therefore assumed to dominate transport control and allow optimum efflux (and uptake) rates. Short-time starvation of carbon and nitrogen adjusts CTP and ATP at levels that cause declining efflux rates. During prolonged starvation both nucleotides fall below their transport-controlling concentrations and thus allow increasing rates of efflux from the still maintained vacuolar pool. Hence, efflux control under nutrient limitation includes an interplay of purine and pyrimidine nucleotides which precisely regulates the release of vacuolar amino acids and enables flexible adjustment to either amino acid saving or cell survival.
Keywords: Vacuole in situ; Amino acid transport; Nucleotide control; Penicillium cyclopium;

We have previously reported that myo-inositol uptake and metabolism is reduced in human fibroblasts derived from patients with ataxia telangiectasia (AT). Treating normal fibroblasts with 10–100 μM wortmannin duplicates some of the phenotypic properties of AT fibroblasts including the decrease in myo-inositol accumulation. In the present study we examined whether treatment of other types of mammalian cells with wortmannin or LY294002 altered myo-inositol uptake. Cultured bovine aorta endothelial cells or 3T3-L1 adipocytes were incubated with either wortmannin or LY294002, and afterwards, myo-inositol uptake and SMIT mRNA levels were determined. Incubating cultured bovine aorta endothelial cells and 3T3-L1 adipocytes with either wortmannin or LY294002 caused a time- and concentration-dependent decrease in myo-inositol accumulation that was independent of changes in SMIT mRNA levels. The effect of wortmannin and LY294002 on myo-inositol accumulation was not due to an increase in myo-inositol secretion. The effect of LY294002 on myo-inositol accumulation was reversible. Furthermore, the LY294002-induced decrease in myo-inositol accumulation was specific since the uptake of serine or choline by cultured bovine aorta endothelial cells and 3T3-L1 adipocytes treated with LY294002 was not significantly decreased. Co-incubation of cultured bovine aorta endothelial cells and 3T3-L1 adipocytes with either wortmannin or LY294002 and hyperosmotic medium caused a significant decrease in the induction of myo-inositol accumulation by hyperosmolarity without significantly affecting the hyperosmotic-induced increase in SMIT mRNA levels. These data suggest that myo-inositol accumulation is regulated post-translationally by wortmannin and LY294002.
Keywords: Wortmannin; LY29002; myo-inositol; Phosphatidylinositol 3-kinase; Endothelial cells; Adipocytes;

Matrix metalloproteinase-2 (MMP-2) has been suggested to play a crucial role in tumor invasion and angiogenesis. In order to understand the mechanisms underlying proMMP-2 activation, we compared the biochemical and cellular events triggered by two potent MMP-2 activators, the lectin concanavalin A (ConA) and the cytoskeleton disrupting agent cytochalasin D (CytoD). Incubation of U87 human glioma cells for 24 h in the presence of ConA or CytoD induced a marked activation of proMMP-2 and this activation was correlated in both cases with an increase in the mRNA levels of MT1-MMP. At the protein level, proMMP-2 activation induced by CytoD or ConA strongly correlated with the appearance of a 43-kDa MT1-MMP proteolytic breakdown product in cell lysates. Interestingly, CytoD also induced a very rapid (2 h) activation of proMMP-2 that was independent of protein synthesis. Under these conditions, CytoD also promoted the rapid proteolytic breakdown of the 63 kDa pro form of MT1-MMP, resulting in the appearance of the 43 kDa MT1-MMP processed form. Overexpression of a recombinant full-length MT1-MMP protein in glioma cells resulted in the activation of proMMP-2 that was correlated with the generation of the 43 kDa fragment of the protein. By contrast, overexpression of the protein in COS-7 cells promoted proMMP-2 activation without inducing the production of the 43 kDa fragment. These results thus suggest that activation of proMMP-2 occurs through both translational and post-translational mechanisms, both involving proteolytic processing of membrane-associated MT1-MMP. This processing of MT1-MMP is, however, not essential to proMMP-2 activation but may represent a regulatory mechanism to control the activity of MT1-MMP.

The inactivation of single-chain urokinase-type plasminogen activator by thrombin on cultured human endothelial cells by Ellen A.M Braat; Annemie Collen; Annie F.H Jie; Jos M Grimbergen; Dingeman C Rijken (351-358).
Single-chain urokinase-type plasminogen activator (scu-PA) is cleaved by thrombin, resulting in an inactive molecule called thrombin-cleaved two-chain urokinase-type plasminogen activator (tcu-PA/T). There is no knowledge about cell-mediated inactivation of scu-PA. We have studied whether scu-PA bound to cultured human umbilical vein endothelial cells (HUVEC) could be inactivated by thrombin. High molecular weight scu-PA was bound to HUVEC and incubated with increasing amounts of thrombin for 30 min at 37°C. Cell-bound urokinase-type plasminogen activator (u-PA) was released and levels of scu-PA, tcu-PA/T and active two-chain u-PA were measured using sensitive bioimmunoassays. Cell-bound scu-PA was efficiently inactivated by thrombin. Fifty percent inactivation of scu-PA occurred at about 0.2 nM thrombin. In the presence of monoclonal anti-urokinase receptor IgG, at least 50% of the binding of scu-PA to HUVEC was inhibited. The relative amount of tcu-PA/T that was generated by thrombin was not affected by the monoclonal antibody. These results indicated that scu-PA bound to HUVEC via the urokinase receptor can be inactivated by thrombin. The efficient inactivation of cell-bound scu-PA suggests that a cofactor for thrombin may be involved, like thrombomodulin or glycosaminoglycans. It is concluded that scu-PA bound to the urokinase receptor on a cell surface can be inactivated by thrombin, which may have profound effects on u-PA-mediated local fibrinolysis and extracellular proteolysis during processes in which thrombin is also involved.
Keywords: Urokinase; scu-PA; Thrombin; HUVEC; Endothelial cells; Urokinase receptor;