BBA - Molecular Cell Research (v.1499, #1-2)

The toxicity of the five methylxanthine derivatives, caffeine, pentoxifylline, A802710, propentofylline and A802715, was determined against the two human melanoma lines, Be11 and MeWo, and against the two human squamous cell carcinoma lines, 4197 and 4451, by vital dye staining assay. Pentoxifylline and A802710 emerge as the least toxic showing TD50 (toxic dose of 50%) levels of 3.0–4.0 mM. Propentofylline and caffeine take an intermediate position. A802715 has a TD50 of 0.9–1.1 mM and is the most toxic. Subtoxic concentrations (<TD50) added after irradiation at maximum expression of the G2/M block show that pentoxifylline and A802710 effectively abrogate the G2/M block, whereas A802715 and propentofylline prolong the G2/M block or remain ineffective depending on the p53 status of the cell line. In p53 wt cells BrdU incorporations show that the irradiation-induced suppression of S-phase entry is marginally enhanced by pentoxifylline but strongly enhanced by propentofylline and A802715. This effect was not seen in p53 mutant cells. Since propentofylline and A802715 prolong the G2/M block and effectively suppress BrdU incorporation these two drugs emerge as antagonists to pentoxifylline, caffeine and A802710. Common structural features of propentofylline and A802715 are a propyl substituent at the N7 position in contrast to pentoxifylline, caffeine and A802710 where the N7 substituent is a methyl group. The results document the effectiveness of four methylxanthines in influencing cell regulation and damage response in human tumor cells.
Keywords: Methylxanthine; G2/M block abrogation; S-phase progression;

The effects of 2-mercapto-1-(β-4-pyridethyl) benzimidazole (MPB), one of the benzimidazole derivatives designed for a nucleic acid analogue, on melanogenesis of murine B16-F10 melanoma cell lines were investigated. MPB (40 μM) induced a striking dendricity in B16 melanoma cells within 12 h and maximal dendricity between 48 and 72 h. The stimulation of melanin synthesis was observed after only 2 days of treatment together with a dose-dependent growth inhibition. Moreover, MPB increased the activity of tyrosinase through the expression of tyrosinase mRNA without increasing the intracellular cyclic AMP content. MPB-induced melanogenesis was inhibited by novel protein kinase A inhibitors, KT-5720 and H-85. These findings indicate that MPB stimulated B16 cells to terminally differentiate and may be a useful drug in studying the regulation of melanogenesis.
Keywords: Melanogenesis; Imidazole derivative; B16 melanoma; Cyclic AMP; α-Melanocyte stimulating hormone; Tyrosinase;

TIMP-1/MMP-9 imbalance in an EBV-immortalized B lymphocyte cellular model: evidence for TIMP-1 multifunctional properties by Philippe Gaudin; Candice Trocmé; Sylvie Berthier; Sylvie Kieffer; Jean Boutonnat; Christophe Lamy; Anny Surla; Jérome Garin; Françoise Morel (19-33).
Tissue inhibitors of metalloproteinases (TIMPs) were initially described as agents controlling metalloproteinase activity. The purpose of this study was to investigate the expression and the roles of TIMP-1 secreted by Epstein–Barr-virus (EBV)-immortalized B lymphocytes. TIMP-1 was isolated from conditioned medium of interleukin (IL)-1β stimulated EBV-B lymphocytes; purified TIMP-1 was identified by mass spectrometry and immunochemistry. TIMP-1-free MMP-9 was quantified after purification by zymography and enzyme-linked immunosorbent assay. EBV-B lymphocyte-secreted TIMP-1 inhibited MMP-9 gelatinolytic activity resulting in decreased B-cell transmigration as measured in vitro. The release of huge amounts of TIMP-1 in proportion to MMP-9 from B lymphocytes after EBV transformation was shown to be correlated with secretion of IL-10 and dependent on culture time. In contrast, there was little TIMP-1 and almost no IL-10 released from native B cells, suggesting a possible IL-10 mediated autocrine regulation mechanism of TIMP-1 synthesis. The MMP-9/TIMP-1 imbalance observed in the culture medium of EBV-B lymphocytes (TIMP-1>MMP-9) and of native B cells (MMP-9>TIMP-1) is suggestive of a new function for TIMP-1. We propose that TIMP-1 acts as a survival factor controlling B-cell growth and apoptosis through an autocrine regulation process involving IL-10 secreted by EBV-B lymphocytes.
Keywords: B lymphocyte; MMP-9; TIMP-1; Growth factor; Apoptosis;

Identification of two focal adhesion targeting sequences in the adapter molecule p130Cas by Mary T Harte; Marlene Macklem; Cheryl L Weidow; J.Thomas Parsons; Amy H Bouton (34-48).
The adapter molecule CAS is localized primarily within focal adhesions in fibroblasts. Because many of the cellular functions attributed to CAS are likely to be dependent on its presence in focal adhesions, this study was undertaken to identify regions of the protein that are involved in its localization. The SH3 domain of CAS, when expressed in isolation from the rest of the protein, was able to target to focal adhesions, whereas a variant containing a point mutation that rendered the SH3 domain unable to associate with FAK remained cytoplasmic. However, in the context of full-length CAS, this mutation did not prevent CAS localization to focal adhesions. Two other variants of CAS that contained deletions of either the SH3 domain alone, or the SH3 domain together with an adjoining proline-rich region, also retained the capacity to localize to focal adhesions. A second focal adhesion targeting region was mapped to the extreme carboxy terminus of CAS. The identification of this second focal adhesion targeting domain in CAS ascribes a previously unknown function to the highly conserved C terminus of CAS. The regulated targeting of CAS to focal adhesions by two independent domains may reflect the important role of CAS within this subcellular compartment.
Keywords: Focal adhesion; p130Cas; Focal adhesion kinase; Src;

κ-Opioid receptor potentiates apoptosis via a phospholipase C pathway in the CNE2 human epithelial tumor cell line by Catherine Tei Mei Diao; Lan Li; See Yan Lau; Tak Ming Wong; Nai Sum Wong (49-62).
The mechanism by which κ-opioid receptor (κor) modulated apoptosis was investigated in CNE2 human epithelial tumor cells. Induction of these cells to undergo apoptosis with staurosporine was associated with a massive increase in intracellular cAMP level. The inhibition of the increase in cAMP partially inhibited apoptosis as evidenced by a reduction of PARP and caspase-3 cleavage. Accordingly, a low but significant level of apoptosis is induced in these cells by the elevation of cAMP through the addition of forskolin and isobutylmethylxanthine. The existence of a cAMP-dependent and a cAMP-independent apoptotic pathway is therefore suggested. Receptor binding studies, RT-PCR experiments and Western blot analysis demonstrated the presence of type 1 κor in the CNE2 cells. Stimulation of κor in these cells resulted in the production of inositol (1,4,5)-trisphosphate, reduction of cAMP level and a marked enhancement of staurosporine-induced apoptosis. The potentiation of apoptosis by κor was prevented by inhibition of phospholipase C but was slightly enhanced by the presence of the active cAMP analogues, 8-CPT-cAMP and dibutyryl-cAMP. These data demonstrate for the first time that the phospholipase C pathway activated by type 1 κor expressed by cancer cells is involved in the potentiation of apoptosis.
Keywords: Opioid; Apoptosis; Inositol; Epithelial; Cyclic adenosine monophosphate;

Acidic ribosomal P-proteins form, in all eukaryotic cells, a lateral protuberance, the so-called ‘stalk’, which is directly involved in translational activity of the ribosomes. In Saccharomyces cerevisiae cells, there are four distinct P-proteins: P1A, P1B, P2A and P2B. In spite of the high level of their structural homology, they are not completely equivalent and may perform different functions. As yet, the protein-protein interactions between yeast P-proteins have not been fully defined. In this paper, the interplay between yeast P-proteins has been investigated by means of a two-hybrid system, chemical cross-linking and gel filtration. The data presented herein show that all P-proteins are able to form homo-oligomeric complexes. By analyzing hetero-interactions, we were able to detect strong interactions between P1A and P2B proteins. Additionally, the pair of P1B and P2A proteins is also able to form a hetero-complex, though at a very low efficiency. All P-proteins are phosphorylated by numerous protein kinases. Using the multifunctional protein kinase CK II, we have shown that incorporation of phosphate into P1A protein can exert its effect on the hetero-oligomerization process, namely by preventing the formation of the hetero-oligomer P1A-P/P2B. These findings are the first to show differences in the oligomerization behavior of the yeast P-proteins; moreover, they emphasize a significant impact of the phosphorylation on the formations of P-protein complex.
Keywords: Protein-protein interaction; Acidic ribosomal P-protein; Phosphorylation;

Glucose deprivation leads to the synthesis of an aberrantly glycosylated (‘alternative’) and inefficiently processed form of the insulin proreceptor in 3T3-L1 adipocytes. To further explore the effect of aberrant (rather than absent) N-linked glycosylation of the insulin receptor, we examined the relationship of processing to function. Our studies show that the alternative form of the proreceptor does not oligomerize nor does it acquire the ability to undergo insulin-sensitive autophosphorylation. This along with an interaction with the glucose-regulated stress protein GRP78/BiP implies inappropriate folding/dimerization and retention in the ER. Glucose refeeding causes the post-translational modification of the alternative form of the proreceptor to a novel ‘intermediate’ form which is independent of new protein synthesis. As little as 100 μM glucose (or mannose) can induce this modification. In vitro digestion of the alternative and intermediate proreceptors with SPC1/furin shows that both the α- and β-subunit domains are glycosylated, albeit aberrantly. This implies that the aberrantly glycosylated proreceptor could serve as a substrate for SPC1 in a physiological setting if the receptor was able to interact with the enzyme in the appropriate compartment (i.e., the trans-Golgi network). Based on inhibitor studies, however, both the alternative and intermediate forms of the proreceptor appear to be primarily targeted to the proteasome for degradation.
Keywords: Insulin receptor processing; Glucose deprivation; 3T3-L1 adipocyte; Alternative glycosylation;

The 110 kDa spindle pole body component, Spc110p, is an essential target of calmodulin in budding yeast. Cells with mutations which reduce calmodulin binding to Spc110p are unable to form a mitotic spindle and die. Here we show that these effects can be overcome either directly by increasing extracellular calcium or calmodulin expression, which reverse the primary spindle defect, or indirectly through increased extracellular osmolarity or high dosage of MID2 or SLG1/HCS77/WSC1 which preserve viability. We propose that overcoming a cell integrity defect associated with the mitotic arrest enables the defective spindle pole bodies to provide sufficient function for proliferation of a large proportion of mutant cells. Our findings demonstrate a role for calcium in the Spc110p-calmodulin interaction in vivo and have important general implications for the interpretation of genetic interactions involving cell integrity genes.
Keywords: Calcium; Calmodulin; Mitosis; Spindle pole body; Signalling; Saccharomyces cerevisiae;

The effect of pulsed electromagnetic fields on the physiologic behaviour of a human astrocytoma cell line by Carlo Aldinucci; Mitri Palmi; Gianpietro Sgaragli; Alberto Benocci; Antonella Meini; Federica Pessina; Gian Paolo Pessina (101-108).
We evaluated the effects of 50 Hz pulsed electromagnetic fields (EMFs) with a peak magnetic field of 3 mT on human astrocytoma cells. Our results clearly demonstrate that, after the cells were exposed to EMFs for 24 h, the basal [Ca2+]i levels increased significantly from 124±51 nM to 200±79 nM. Pretreatment of the cells with 1.2 μM substance P increased the [Ca2+]i to 555±278 nM, while EMF exposure caused a significant drop in [Ca2+]i to 327±146 nM. The overall effect of EMFs probably depends on the prevailing Ca2+ conditions of the cells. After exposure, the proliferative responses of both normal and substance P-pretreated cells increased slightly from 1.03 to 1.07 and 1.04 to 1.06, respectively. U-373 MG cells spontaneously released about 10 pg/ml of interleukin-6 which was significantly increased after the addition of substance P. Moreover, immediately after EMF exposure and 24 h thereafter, the interleukin-6 levels were more elevated (about 40%) than in controls. On the whole, our data suggest that, by changing the properties of cell membranes, EMFs can influence Ca2+ transport processes and hence Ca2+ homeostasis. The increased levels of interleukin-6 after 24 h of EMF exposure may confirm the complex connection between Ca2+ levels, substance P and the cytokine network.
Keywords: Intracellular calcium; Cell proliferation; Substance P; Caffeine; Cytokine; Interleukin-6;

The effects of progesterone and 17β-estradiol on Ca2+ signaling in in situ endothelial cells were investigated using front-surface fluorometry of fura-2-loaded strips of porcine aortic valve. Progesterone inhibited the thapsigargin-induced sustained [Ca2+]i elevation (IC50=33.9 μM, n=4), while 17β-estradiol added a transient [Ca2+]i elevation. Progesterone and 17β-estradiol had no significant effect on the thapsigargin-induced [Ca2+]i elevations in the absence of extracellular Ca2+. A Mn2+-induced decline of fluorescent intensity at 360 nm excitation was accelerated by thapsigargin. This acceleration was completely reversed by progesterone, but not by 17β-estradiol. Progesterone inhibited, and 17β-estradiol enhanced the endothelin-1 (ET-1)-induced [Ca2+]i elevation, while both had no effect on the ET-1-induced Ca2+ release observed in the absence of extracellular Ca2+ or in the pertussis toxin-treated strips. Progesterone and 17β-estradiol thus had different effects on Ca2+ signaling, especially on Ca2+ influx, in endothelial cells.
Keywords: Progesterone; 17β-Estradiol; Endothelial cell; Thapsigargin; Calcium influx; Manganese quenching;

Transrepression of NF-κB is not required for glucocorticoid-mediated protection of TNF-α-induced apoptosis on fibroblasts by Mónica A. Costas; Lionel Müller Igaz; Florian Holsboer; Eduardo Arzt (122-129).
The cellular resistance to tumor necrosis factor (TNF) of most cell types has been attributed to both a protective pathway induced by this cytokine and the preexistence of protective factors in the target cell. NF-κB has been postulated as one of the principal factors involved in antiapoptotic gene expression control on TNF-resistant cells. We have previously shown that glucocorticoids protect the naturally TNF-sensitive L-929 cells from apoptosis. Here we analyze the role of NF-κB and glucocorticoids on TNF-induced apoptosis in L-929 cells. We found that inhibition of NF-κB enhanced the sensitivity to TNF-induced apoptosis. Glucocorticoids inhibited NF-κB transactivation via IκB induction. Moreover, glucocorticoids protected from TNF-induced apoptosis even when NF-κB activity was inhibited by stable or transient expression of the superrepressor IκB. These results demonstrate that although glucocorticoids inhibit NF-κB transactivation in these cells, this is not required for their protection from TNF-induced apoptosis.
Keywords: Apoptosis; Glucocorticoid; Glucocorticoid receptor; Nuclear factor-κB; Tumor necrosis factor;

Expression and intracellular localization of leptin receptor long isoform-GFP chimera by Anna Lundin; Helena Rondahl; Erik Walum; Mona Wilcke (130-138).
The leptin receptor (OBR) and its ligand leptin (OB) are key players in the regulation of body weight. The OBR is a member of the class I cytokine receptor family and is alternatively spliced into at least six different isoforms. The multiple forms are identical in their extracellular and transmembrane regions but differ in lengths. The two predominant isoforms include a long form (OBRl) with an intracellular domain of 303 amino acids and a shorter form (OBRs) with an intracellular domain of 34 amino acids. We have constructed a recombinant OBRl chimera with the green fluorescent protein (GFP) by fusing GFP to the C-terminus of the OBRl. The OBRl-GFP chimera was transiently transfected and expressed in SHSY5Y and HEK293 cells. In a STAT-Luciferase assay we show that the GFP moiety in this chimera did not affect the signalling capacity of OBRl-GFP. In both SHSY5Y and HEK293 cells transfected with OBRl-GFP, a predominant intracellular green OBRl-GFP fluorescence was detected in vesicles also positive for internalized fluorophore conjugated leptin. We also found that treatment with the lysosomotropic reagent monensin did not relocalize OBRl-GFP together with the human transferrin receptor in recycling endosomes, indicating OBRl-GFP not to participate in this pathway. In biotinylation-streptavidin pulse chase experiments, using antibodies raised against GFP and OBR, we observed that the rate of early appearance of OBRs at the cell surface, upon leptin stimulation, was faster than that found for OBRl-GFP. Taken together, our results provide novel data concerning the intracellular trafficking of the two different isoforms of the leptin receptor.
Keywords: Obesity; Leptin; Leptin receptor; Endocytosis; Green fluorescent protein;

Efficient transformation of Dictyostelium discoideum with a particle inflow gun by Birgit Wetterauer; Klaus Salger; Petra Demel; Hans-Ulrich Koop (139-143).
We report experiments to transform Dictyostelium discoideum using a simple home-made particle gun. Stable transformants were obtained at frequencies of up to 2500 clones/μg DNA. This is five times more than we achieve with the same vector using electroporation protocols. We also show that the particle inflow gun can be used for analysis of developmentally regulated gene expression in a transient assay.
Keywords: Transformation; Biolistics; Particle gun; Extrachromosomal vector; Ddp1; (Dictyostelium);

S-Allylmercaptoglutathione: the reaction product of allicin with glutathione possesses SH-modifying and antioxidant properties by Aharon Rabinkov; Talia Miron; David Mirelman; Meir Wilchek; Sabina Glozman; Ephraim Yavin; Lev Weiner (144-153).
The reaction between allicin (diallylthiosulfinate), the active component of garlic and reduced glutathione was investigated. The product of this reaction, mixed disulfide S-allylmercaptoglutathione (GSSA) was separated by high performance liquid chromatography and identified by 1H and 13C nuclear magnetic resonance and mass spectroscopy. The reaction is fast (with an apparent bimolecular reaction rate constant of 3.0 M−1 s−1). It is pH-dependent, which reveals a direct correlation to the actual concentration of mercaptide ion (GS). Both GSSA and S-allylmercaptocysteine (prepared from allicin and cysteine) reacted with SH-containing enzymes, papain and alcohol dehydrogenase from Thermoanaerobium brockii yielding the corresponding S-allylmercapto proteins, and caused inactivation of the enzymes. The activity was restored with dithiothreitol or 2-mercaptoethanol. In addition, GSSA also exhibited high antioxidant properties. It showed significant inhibition of the reaction between OH radicals and the spin trap 5,5′-dimethyl-1-pyroline N-oxide in the Fenton system as well as in the UV photolysis of H2O2. In ex vivo experiments done with fetal brain slices under iron-induced oxidative stress, GSSA significantly lowered the production levels of lipid peroxides. The similar activity of GSSA and allicin as SH-modifiers and antioxidants suggests that the thioallyl moiety has a key role in the biological activity of allicin and its derivatives.
Keywords: Garlic; Allicin; S-Allylmercaptoglutathione; S-Allylmercaptocysteine; Enzyme inactivation; Antioxidant activity;

A ubiquitination factor, NosA, is essential for cellular differentiation in Dictyostelium discoideum. In the absence of nosA, development is blocked, resulting in a developmental arrest at the tight-aggregate stage, when cells differentiate into two precursor cell types, prespore and prestalk cells. Development is restored when a second gene, encoding the ubiquitin-like protein SonA, is inactivated in nosA-mutant cells. SonA has homology over its entire length to Dsk2 from Saccharomyces cerevisiae, a ubiquitin-like protein that is involved in the assembly of the spindle pole body. Dsk2 and SonA are both stable proteins that do not seem to be subjected to degradation via the ubiquitin pathway. SonA does not become ubiquitinated and the intracellular levels of SonA are not affected by the absence of NosA. The high degree of suppression suggests that SonA rescues most or all of the defects caused by the absence of nosA. We propose that NosA and SonA act in concert to control the activity of a developmental regulator that must be deactivated for cells to cross a developmental boundary.
Keywords: Proteolysis; Parkin; dsk2; Slime mold; Ubiquitin-like domain;

We have cloned a cDNA encoding a catalytic subunit of calcineurin (CnA) expressed in Xenopus oocytes. The deduced amino acid sequence indicates 96.3% and 96.8% identities with the mouse and human CnAα isoforms, respectively. Xenopus CnA (XCnA) RNA and protein are expressed as maternal and throughout development. Recombinant XCnA protein interacted with calmodulin in the presence of Ca2+. Deletion of calmodulin binding domain and auto-inhibitory domain revealed calcium independent phosphatase activity, thereby showing that XCnA is likely to be modulated by both calmodulin and calcium.
Keywords: Calcineurin; Xenopus; cDNA cloning; Calcium dependent; Alternative splicing;