Expression and function of protein phosphatase PP2A in malignant testicular germ cell tumours.

Testicular germ cell tumours (TGCT) represent the most common malignancy in young males. We reported previously that two prototype members of the mitogen-activated protein kinase (MAPK) family, the MAPK ERK kinase (MEK) and extracellular signal-regulated kinase (ERK), are inactive in malignant testicular germ cells and become active after drug stimulation, leading to apoptosis of tumour cells. In this study, we asked whether the protein phosphatase PP2A, a known inhibitor of the MEK-ERK pathway, participates in the proliferation and/or apoptosis of primary TGCT (n = 48) as well as two TGCT cell lines (NTERA and NCCIT). Quantitative RT-PCR, immunohistochemistry, western blot analyses and phosphatase assay indicate that primary TGCT as well as TGCT cell lines express PP2A and that PP2A is active in TGCT cell lines. The inhibition of PP2A by application of two PP2A inhibitors, cantharidic acid (CA) and okadaic acid (OA), results in a significant increase in caspase-3-mediated apoptosis of TGCT cell lines. Thereby, PP2A inhibition was accompanied by phosphorylation and activation of MEK and ERK. Functional assays using the MEK inhibitor PD98059 demonstrated that the phosphorylation of MEK and ERK was required for the induction of caspase-3-mediated apoptosis of malignant germ cells. Thus, our data suggest that inhibition of PP2A mediates its apoptosis-inducing effect on TGCT through activation of the MEK-ERK signalling pathway that leads to caspase-3-mediated apoptosis of tumour cells. In addition our results support previous observations that PP2A exerts an anti-apoptotic effect on malignant tumour cells.

Apoptosis by 6-O-palmitoyl-L-ascorbic acid coincides with JNK-phosphorylation and inhibition of Mg2+-dependent phosphatase activity.

6-O-Palmitoyl ascorbic acid (PAA) has recently been used as a substitute for ascorbic acid because of its greater potency as an antioxidant. In detailed concentration response studies distinct cytotoxic effects of PAA at concentrations exceeding 100 microM were reported. Here we examined and further characterized this cytotoxicity. While ascorbic acid was tolerated well up to millimolar concentrations, PAA revealed an LC50 between 125 and 150 microM in rat GH3 tumor cells. Morphological and biochemical observations suggested the induction of apoptosis at concentrations exceeding 125 microM with a prominent activation of caspase 3 at 250 microM after 4 hr. A subsequent pronounced fragmentation of DNA (DNA-ladder) was detected after 6 hr and was further enhanced after 12 hr. The activation of caspases and the cleavage of its substrate PARP was preceded by a distinct increase in the phosphorylation of stress activated JNK-kinases. This observation suggested that the agent affected signal transduction mechanisms regulating protein phosphorylation at serine/threonine residues in the cell. No effect of PAA on protein phosphatase 2A (PP2A)-like activity was observed while magnesium-dependent protein phosphatase activity, presumably PP2C, was inhibited concentration-dependently up to 75% at the respective concentrations. Thus, the cytotoxic, pro-apoptotic effect of PAA might be related to the inhibition of PP2C and the activation of JNK.

A fast and sensitive technique to study the kinetics and the concentration dependencies of DNA fragmentation during drug-induced apoptosis.

Apoptotic cell death with its characteristic coordinated cellular breakdown can be triggered by cytotoxic drugs. One prominent feature that differentiates apoptotic from necrotic cell death is the caspase-mediated activation of an endonuclease that internucleosomally cleaves DNA resulting in the so-called apoptotic DNA ladder. Here we report a new rapid, sensitive and inexpensive column separation technique to study drug-induced DNA fragmentation from 10(6) or less cells. This technique, which is based on a modified plasmid spin column kit, avoids the use of hazardous chemicals. With this procedure and subsequent densitometric analysis it was possible to study the concentration dependencies and the kinetics of drug-induced DNA fragmentation. The applicability of this technique is shown for okadaic acid- and cantharidic-acid-treated pituitary GH(3) cells as well as highly okadaic-acid-resistant sublines. These studies allowed us to compare as well as to differentiate the effects and potencies of these structurally different but functionally quite similar inhibitors of ser/thr phosphatases 1 and 2A.

Michael adducts of ascorbic acid as inhibitors of protein phosphatase 2A and inducers of apoptosis.

Michael adducts of ascorbic acid with alpha,beta-unsaturated carbonyl compounds have been shown to be potent inhibitors of protein phosphatase 1 (PP1) without affecting cell viability at the respective concentrations. Here we were able to show that higher concentrations can partially inhibit PP2A activity and concomitantly induce apoptotic cell death. A nitrostyrene adduct of ascorbic acid proved to be a more potent and effective inhibitor of PP2A as well as a stronger inducer of apoptosis. These adducts only slightly lost their cytotoxic potential in multidrug resistant cells that were 10-fold less sensitive to apoptosis induction by okadaic acid and vinblastine.

Hamster pancreatic beta cell lines with altered sensitivity towards apoptotic signalling by phosphatase inhibitors.

Specific inhibitors of serine/threonine phosphatases like okadaic acid can induce apoptotic cell death in the pancreatic beta cell line HIT. Cultivation in stepwise increased concentrations of okadaic acid enabled the isolation of HIT100R cells which proliferate at 100 nM okadaic acid (8 - 10 times the initially lethal concentration). These two cell lines were used to characterize the events triggered by okadaic acid that led to apoptosis. Biochemical markers, e.g. cytochrome c release from mitochondria and increase of caspase-3-like activity, revealed that induction of apoptosis by 100 nM okadaic acid in parental HIT cells started with the release of cytochrome c. In HIT100R cells 500 nM okadaic acid were necessary to induce alterations comparable to those observed with 100 nM okadaic acid in non-resistant HIT cells. In contrast to okadaic acid, the potency of the structurally different phosphatase inhibitor cantharidic acid to induce cytochrome c release, increase of caspase-3-like activity and DNA fragmentation was comparable in HIT and HIT100R cells. Thus, no cross-resistance between these phosphatase inhibitors seemed to exist. Phosphatase activity in extracts from HIT and HIT100R cells did not differ in its total amount or in its sensitivity for okadaic acid. Since higher concentrations of okadaic acid were needed to induce apoptosis in HIT100R cells, a compromised intracellular accumulation of the toxin appeared likely. Functional and structural analysis revealed that this was achieved by the development of the multidrug resistance phenotype in HIT100R cells. The underlying mechanism appeared to be the enhanced expression of the pgp1 but not the pgp2 gene.

Contribution of mdr1b-type P-glycoprotein to okadaic acid resistance in rat pituitary GH3 cells.

Okadaic acid as well as other, structurally different, inhibitors of serine/threonine phosphatases 1 and 2A induce apoptosis in pituitary GH3 cells. Incubation with stepwise raised concentrations of okadaic acid resulted in the isolation of cells that were increasingly less sensitive to the cytotoxic effect of this agent. After about 18 months cells were selected that survived at 300 nM okadaic acid, which is about 30 times the initially lethal concentration. This study revealed that a major pharmacokinetic mechanism underlying cell survival was the development of a P-glycoprotein-mediated multidrug resistance (MDR) phenotype. The increase in mRNA levels of the mdr1b P-glycoprotein isoform correlated with the extent of drug resistance. Functional assays revealed that increasing drug resistance was paralleled by a decreased accumulation of rhodamine 123, a fluorescent dye which is a substrate of mdr1-mediated efflux activity. Resistance could be abolished by structurally different chemosensitizers of P-glycoprotein function like verapamil and reserpine but not by the leukotriene receptor antagonist MK571 which is a modulator of the multidrug resistance-associated protein (MRP). Okadaic acid resistance included cross-resistance to other cytotoxic agents that are substrates of mdr1-type P-glycoproteins, like doxorubicin and actinomycin D, but not to non-substrates of mdr1, e.g. cytosine arabinoside. Thus, functional as well as biochemical features support the conclusion that okadaic acid is a substrate of the mdr1-mediated efflux activity in rat pituitary GH3 cells. Maintenance of resistance after withdrawal of okadaic acid as well as metaphase spreads of 100 nM okadaic acid-resistant cells suggested a stable MDR genotype without indications for the occurrence of extrachromosomal amplifications, e.g. double minute chromosomes.

Doxorubicin-resistant LoVo adenocarcinoma cells display resistance to apoptosis induction by some but not all inhibitors of ser/thr phosphatases 1 and 2A.

LoVo adenocarcinoma cells are fairly sensitive to cytostatic drugs, e.g. doxorubicin, but can develop drug resistance by expression of a P-glycoprotein-mediated MDR1 phenotype. LoVo cells respond with apoptosis to nanomolar concentrations of okadaic acid and micromolar concentrations of cantharidic acid. Interestingly, LoVoDx cells which had become about 10-fold less sensitive to doxorubicin by incubation in increasing concentrations of this cytostatic drug were also less sensitive to the toxicity of okadaic acid. Resistance to both agents was lost or significantly reduced by incubation in drug-free medium for about 4 months. On the other hand, LoVoDx cells did not lose responsiveness to the structurally different phosphatase inhibitor cantharidic acid but were about twofold more sensitive to the cytotoxic effect of this agent. Thus, MDR expression protects LoVo cells from the toxicity of phosphatase inhibitors that presumably are substrates of the P-glycoprotein, e.g. okadaic acid and its derivatives but not cantharidic acid, despite the fact that both agents are potent inducers of apoptotic cell death via ser/thr phosphatase inhibition.

Phosphatase inhibitors induce defective hormone secretion in insulin-secreting cells and entry into apoptosis.

A long-term (> or =24 h) exposure of insulin-secreting HIT T15 cells to the phosphatase inhibitor, okadaic acid (OA), at concentrations inhibiting serine/threonine phosphatases 1 (PP1) and 2A (PP2A) reduced proliferation and insulin secretion. The reduced proliferation was related to the induction of apoptosis as evidenced by morphological criteria and the occurrence of internucleosomal DNA fragmentation after 15 h in 50 nM OA. The compromised insulin secretion was not simply a consequence of a lowered hormone content and cell growth, but comprised also a complete suppression of secretion stimulated by K+ depolarisation and forskolin. K+ depolarisation of HIT cells cultured for 24 h in 50 nM OA resulted in a nearly unimpaired influx of Ca2+, but did not induce secretion. These observations suggest that the secretory defect may be localised distal to Ca2+ influx in stimulus secretion coupling of insulin-secreting cells.

Characterization of two pituitary GH3 cell sublines partially resistant to apoptosis induction by okadaic acid.

Pituitary GH3 cells die by apoptosis when treated with okadaic acid, a specific inhibitor of ser/thr phosphatases. Incubations starting at concentrations of 5 and 12.5 nM followed by stepwise rises resulted in two populations (the S1 and S2 sublines) that proliferated at initially lethal 30 nM. Cells were partially resistant to higher concentrations of okadaic acid and its derivative methyl okadaate. Toxicity of the structurally distinct inhibitors cantharidic acid and calyculin A was differently affected in the two resistant lines. The enhanced expression of the P-glycoprotein was one mechanism of resistance in S1 and S2. Resistance was reversed completely (S1) or partially (S2) by the addition of verapamil. In addition, phosphatase activity, presumably PP2A, was increased in S2. Therefore, pharmacokinetic and pharmacodynamic mechanisms can protect pituitary GH3 cells from apoptotic cell death by okadaic acid.

Apoptosis induction by inhibitors of Ser/Thr phosphatases 1 and 2A is associated with transglutaminase activation in two different human epithelial tumour lines.

Two epithelial tumour lines, HeLa and KB, were treated with okadaic acid and calyculin A, specific inhibitors of Ser/Thr phosphatases (PP), esp. PP1 and PP2A. Morphological criteria, analysis of DNA fragmentation and studies of membrane integrity revealed that both agents concentration- and time-dependently induced apoptosis at nanomolar concentrations which in these cells was associated with the stimulation of a transglutaminase activity. Since a non-functional derivative of okadaic acid did not affect cell viability apoptosis was apparently related to the inhibition of PP1 and PP2A. Membrane damage marker activity was delayed by at least 24 h when compared to nuclear alterations.

Inhibitors of ser/thr phosphatases 1 and 2A induce apoptosis in pituitary GH3 cells.

Two structurally different inhibitors of ser/thr phosphatases 1 and 2A, okadaic acid and calyculin A, time- and concentration-dependently stimulated and inhibited cell-specific function (hormone gene expression) in pituitary GH3 cells. The negative effect was associated with the appearance of apoptotic cell death. Nanomolar concentrations of both agents produced the characteristic morphological alterations and a DNA fragmentation ladder. Calyculin A treatment resulted in comparable changes with 10fold lower concentrations than okadaic acid. Observations with derivatives of okadaic acid with no or lower phosphatase inhibitory potency supported the conclusion that apoptosis induction is related to inhibition of ser/thr phosphatases, presumably types 1 and 2A. Membrane damage as measured by lactate dehydrogenase liberation into medium was significantly lower in apoptotic vs. necrotic cells. DNA fragmentation could be reduced by the addition of zinc but not by removal of extracellular calcium with EGTA. Apoptotic changes were reduced by the concomitant activation of protein kinase A by a membrane permeable cAMP analogue. Incubation of cells for 4 months in successively increased concentrations of okadaic acid resulted in a population that proliferated at the initially lethal concentration of 30 nM.

Thyrotropin (TSH) beta-subunit gene expression--an example for the complex regulation of pituitary hormone genes.

Synthesis of pituitary hormones was shown to be efficiently regulated at the transcriptional level. The specialized function of the five cell types in the anterior pituitary is controlled by ubiquitous as well as cell-specific transcription factors. Pit-1 is such a cell-specific regulator found only in lacto-, somato- and thyrotropes which could be shown to be essential for basal expression of growth hormone (GH) and prolactin (Prl) genes and the regulated expression of Prl and thyrotropin (TSH) beta-subunit genes. Identification of distinct binding sites for transcription factors and some of the mechanisms of transcriptional control shed light on the complex regulation of pituitary hormone gene expression which is exemplified for the TSH beta gene. The control of basal as well as positively and negatively regulated expression of some pituitary hormone genes becomes fairly well understood by the investigation of the role of Pit-1. Identification of different mutations in the human pit-1 gene supported the role of this protein for combined pituitary hormone deficiency (CPHD) characterized by the deficiency of GH, prolactin and TSH.

AP-1 antagonizes thyroid hormone receptor action on the thyrotropin beta-subunit gene.

Thyrotropin-releasing hormone (TRH) stimulates and thyroid hormone (T3) inhibits transcription of the thyrotropin beta-subunit gene (TSH-beta). The first exon contains DNA sequences necessary for both responses and binds both AP-1 and thyroid hormone receptor (T3R). T3 did not inhibit TSH-beta gene expression in a T3R-deficient cell line. Transfection of a T3R expression vector, however, resulted in a 70% inhibition of expression by T3, which was abolished by cotransfection of c-jun and c-fos expression vectors. Mutations surrounding the transcription initiation site and DNA binding studies demonstrate both a functional and structural interaction between c-jun and T3R. Thus, TRH, acting through AP-1, may alter the set point and magnitude of thyroid hormone negative feedback of the TSH-beta gene through an interaction between AP-1 and T3R. Other regulatory pathways acting through AP-1 may alter thyroid hormone action in man.

Hormonal regulation of the thyrotropin beta-subunit gene by phosphorylation of the pituitary-specific transcription factor Pit-1.

The pituitary-specific transcription factor Pit-1 is a cell-specific activator of prolactin and growth hormone gene transcription in the anterior pituitary. Pit-1 has also been shown to mediate both thyrotropin-releasing hormone (TRH) and cAMP stimulation of the prolactin and thyrotropin beta-subunit (TSH beta) genes. The molecular mechanism by which Pit-1 mediates these stimulatory effects remains unclear. At least three Pit-1-binding elements within the TSH beta gene mediate responsiveness to TRH and cAMP. The present studies were designed to test the hypothesis that phosphorylation is an important modulator of Pit-1 interaction with the TSH beta gene. TSH beta elements bind less well to nonphosphorylated Pit-1 than to phosphorylated Pit-1 and are weak activators of gene expression, unlike high-affinity Pit-1 binding sites in the prolactin and growth hormone genes. Phosphorylation by protein kinase A or C enhances Pit-1 binding to TSH beta elements 3- to 8-fold. Conversely, phosphorylation generally reduces binding of Pit-1 to elements within the prolactin and growth hormone genes. A variation within the consensus sequence for Pit-1 binding in TSH beta gene elements [A(A/T)(A/T)AATNCAT in the TSH beta gene versus A(A/T)(A/T)TATNCAT in the prolactin and growth hormone genes] could explain these differences. These elements may limit basal activation of the TSH beta gene by binding less well to nonphosphorylated Pit-1 while conferring hormonal stimulation through enhanced binding of phosphorylated Pit-1.

Role of a pituitary-specific transcription factor (pit-1/GHF-1) or a closely related protein in cAMP regulation of human thyrotropin-beta subunit gene expression.

cAMP regulation of the human thyrotropin-beta (TSH beta) gene cAMP was studied in two heterologous cell lines, a human embryonal kidney cell line (293) and a rat pituitary cell line (GH3). In 293 cells, human TSH beta gene expression was not stimulated by the adenylate cyclase activator forskolin or the cAMP analogue 8-bromo-cAMP (8-Br-cAMP). On the other hand, these agents induced human TSH beta gene expression 4-12-fold in GH3 cells. Deletion analysis demonstrated that the regions from +3 to +8 bp and from -128 to -61 bp were both necessary for cAMP stimulation. The latter region contains three DNA sequences homologous to a pituitary-specific transcription factor, Pit-1/GHF-1, DNA-binding site. Gel-mobility assays demonstrated that a radiolabeled human TSH beta probe (-128 to -61 bp) formed five specific DNA-protein complexes with mouse thyrotropic tumor (MTT) nuclear extract and two specific complexes with in vitro translated Pit-1/GHF-1. Four of the five MTT complexes and both in vitro Pit-1/GHF-1 complexes were reduced or eliminated by excess of an unlabeled Pit-1/GHF-1 DNA-binding site from the rat growth hormone gene, but not a mutated version of the same DNA fragment, suggesting that Pit-1/GHF-1 or a closely related thyrotroph protein binds to these DNA sequences. In 293 cells, co-transfection of an expression vector containing the Pit-1/GHF-1 cDNA restored cAMP-responsiveness to the human TSH beta promoter (5.2- and 6.6-fold maximal stimulation by 8-Br-cAMP and forskolin, respectively) but not the herpes virus thymidine kinase promoter (1.2-fold maximal stimulation by either agent). Thus we conclude that the human TSH beta gene is positively regulated by cAMP in GH3 but not 293 cells. Since the human TSH beta gene contains at least one high-affinity binding site for Pit-1/GHF-1 in a region necessary for cAMP stimulation and cAMP stimulation could be restored to the human TSH beta promoter in a previously nonresponsive cell line by the addition of Pit-1/GHF-1, this suggests that Pit-1/GHF-1, or a closely related protein in the thyrotroph, may be a trans-acting factor for cAMP stimulation of the TSH beta gene.

Thyrotropin-releasing hormone regulation of human TSHB expression: role of a pituitary-specific transcription factor (Pit-1/GHF-1) and potential interaction with a thyroid hormone-inhibitory element.

Regulation of human thyrotropin beta subunit gene (TSHB) expression by thyrotropin-releasing hormone (TRH) was examined in a clonal rat pituitary-cell line (GH3). Transient expression studies were done with various 5'-flanking DNA sequences of TSHB coupled to reporter gene chloramphenicol acetyltransferase. Deletion analysis defined two discrete regions (-128 to -92 base pairs and -28 to +8 base pairs) that each mediated an approximately 2-fold TRH induction. The upstream site contains a DNA sequence with close homology to the DNA-binding site for a pituitary-specific transcriptional factor Pit-1/GHF-1. DNase I footprinting analysis of mouse thyrotropic tumor extract as well as DNA-transfection studies using an expression vector containing an N-terminal deletion of Pit-1/GHF-1 cDNA suggest that Pit-1/GHF-1 or a closely related protein in the thyrotroph mediates TRH responsiveness of this gene. In addition, the downstream site overlaps with the recently characterized thyroid hormone-inhibitory element of TSHB. In fact, deletion of DNA sequences important in thyroid hormone-receptor binding (c-erbAB/c-ERBA2) from +3 to +8 base pairs, significantly reduced (30%) TRH responsiveness. The location of a TRH-stimulatory element near a thyroid hormone-inhibitory element may allow for fine control of TSHB expression in vivo.

Methylxanthines inhibit glucose transport in rat adipocytes by two independent mechanisms.

In summary, this study characterized the biphasic inhibition of fat cell glucose transport by the lipolytic agents caffeine and theophylline. Like the lipolytic drug forskolin, both methylxanthines produced an immediate inhibition of glucose transport that was not seen with 8-phenyltheophylline, a pure adenosine receptor antagonist. The immediate inhibition was therefore not mediated by the adenosine receptor antagonism but seems to be due to a direct interaction with the hexose transporter. This conclusion is supported by the immediate onset of the inhibition and additionally by the interference of theophylline and caffeine with the binding of cytochalasin B, a ligand of the glucose transporter that binds to an intracellular site of the transporter molecule. In addition, a second, delayed inhibitory effect of theophylline and caffeine on glucose transport was observed. This portion shared many aspects of the inhibitory effect of lipolytic hormones. It developed over a period of about 5 min and was antagonized by the simultaneous addition of the antilipolytic hormone PGE2. This component of transport inhibition could be attributed to the antagonistic effect of methylxanthines at the fat cell A1-adenosine receptor since it was also seen with 8-phenyltheophylline. This conclusion is further supported by data showing that the removal of endogenous adenosine with adenosine deaminase resulted in a comparable 25-30% inhibition of insulin-stimulated glucose transport. In addition, the time course of glucose transport inhibition by the subsequent addition of adenosine deaminase is similar to that of the delayed portion of the inhibition seen with theophylline and caffeine. Both treatments produced their maximal inhibition after 5 min. In conclusion, the methylxanthines theophylline and caffeine inhibit glucose transport by a combination of two different modes of action. The immediate major component is mediated via a direct interaction with the hexose transporter whereas the delayed component involves adenosine receptor antagonism and thereby the interaction with G-proteins.

Pre-translational and post-translational regulation of TSH: relationship to bioactivity.

We are interested in the mechanisms by which endocrine and developmental factors regulate TSH synthesis at both pre-translational and post-translational levels. Thyroid hormone profoundly decreases transcription of the TSH beta gene, while TRH and agents modifying cyclic AMP increase transcription. To elucidate the molecular mechanisms underlying these effects, human embryonal kidney cells were transfected with constructs of the human TSH-beta gene fused to the chloramphenicol acetyl transferase gene. The first exon of human TSH-beta contains an element that increases basal expression and mediates T3-induced gene repression, probably through a direct interaction with c-erbA beta. In contrast, TRH and agents modifying cyclic AMP mediate increased transcription of TSH-beta through interacting with upstream regulatory elements. Thyroid hormone, TRH and developmental factors also regulate the branching pattern and relative sialylation of TSH carbohydrate chains, which may affect TSH action in vitro and in vivo.

Thyroid hormone inhibition of human thyrotropin beta-subunit gene expression is mediated by a cis-acting element located in the first exon.

Thyroid hormone regulation of the human thyrotropin beta-subunit gene (TSH beta) was examined in a human embryonal cell line (293). Transient expression studies were performed with chimeric plasmids containing the reporter gene, chloramphenicol acetyltransferase. Sequences in the first exon between +9 and +37 base pairs (bp) enhanced gene expression from the human TSH beta promoter in the absence of thyroid hormone as well as mediated a concentration-dependent triiodothyronine (L-T3) decrease in gene expression. Thyroid hormone inhibition of expression was also conferred to the herpes simplex virus thymidine kinase promoter by inserting +3 to +37 bp of the human TSH beta gene downstream from the start of transcription. Primer extension analysis of RNA from transfected cell cultures revealed accurate transcription initiation in only those constructs which contained sequences between +9 and +37 bp. Moreover, RNA analysis confirmed that L-T3 inhibition of chloramphenicol acetyltransferase activity from chimeric pTSH beta CAT constructs occurred at a pretranslational level. In addition, a nuclear thyroid hormone receptor, c-erbA-beta, bound to this region in an avidin-biotin DNA binding assay. These data suggest that L-T3, bound to its receptor, may inhibit human TSH beta expression by interfering with an element that functions to enhance gene expression.