Control of specific gene expression examined in synchronized mammalian cells

Modulation of glucocorticoid receptor function via phosphorylation

The glucocorticoid receptor (GR) is phosphorylated at multiple serine residues in a hormone-dependent manner. It has been suggested that GR phosphorylation affects turnover, subcellular trafficking, or the transcriptional regulatory functions of the receptor, yet the contribution of individual GR phosphorylation sites to the modulation of GR activity remains enigmatic. This review critically evaluates the literature on GR phosphorylation and presents more recent work on the mechanism of GR phosphorylation from studies using antibodies that recognize GR only when it is phosphorylated. In addition, we present support for the notion that GR phosphorylation modifies protein-protein interactions, which can stabilize the hypophosphorylated form of the receptor in the absence of ligand, as well as facilitate transcriptional activation by the hyperphosphorylation of GR via cofactor recruitment upon ligand binding. Finally, we propose that GR phosphorylation also participates in the nongenomic activation of cytoplasmic signaling pathways evoked by GR. Thus, GR phosphorylation is a versatile mechanism for modulating and integrating multiple receptor functions.

Regulation of tyrosine aminotransferase gene expression by glucocorticoids in quiescent and regenerating liver

Following 70% hepatectomy, the induction of tyrosine amino-transferase mRNA by glucocorticoids was marginal at 1.5 h, significantly impaired between 3 and 8 h and, at 16 h post-hepatectomy, reached a value approx. 5-fold the basal level, similar to the level observed in quiescent liver. The fold induction of the mRNA was accounted for by a similar fold activation of transcription of the gene by glucocorticoids in regenerating but not in quiescent liver; in the latter, activation of transcription was marginal in spite of glucocorticoid-induced hypersensitivity to cleavage by DNase I at the glucocorticoid-dependent enhancer of the gene. The possibility that in quiescent liver glucocorticoids act at a transcriptional step beyond initiation, increasing the rate of elongation or overcoming a blockage in elongation, was excluded. However, a similar fold induction was determined for total and nuclear tyrosine aminotransferase mRNA in the presence of glucocorticoids, suggesting that in quiescent liver glucocorticoids promote efficient maturation of the tyrosine aminotransferase primary transcript. Thus a glucocorticoid-induced nuclear post-transcriptional up-regulation apparently compensates for impaired activation of transcription of the tyrosine aminotransferase gene by glucocorticoids in quiescent liver.

Regulation of the synthesis of tyrosine aminotransferase: the relationship to mRNATAT

The activity of the hepatic enzyme tyrosine aminotransferase (TAT) is the sum of many diverse regulatory factors. These include the developmental stage of the animal, the hormonal and nutritional environment of the animal (or tissue culture cell), other extrinsic and intrinsic regulatory cycles and factors (including cytoplasmic substances), and chromatin structure. Although TAT is subject to a number of post-translational modifications, alterations in catalytic activity always parallel changes in enzyme amount. In a few instances this is due to a selective change in TAT degradation, but most are due to changes in the rate of aminotransferase synthesis. Recent studies have shown that TAT synthesis is generally directly correlated with the activity, and presumably amount, of the mRNA that codes for tyrosine aminotransferase.

Laser flow cytofluorometric analysis of HTC cells synchronized with hydroxyurea, nocodazole and aphidicolin

The technique of laser flow cytofluorometry has been used to monitor the arrival in G1 and the subsequent progression through the cell cycle of HTC cells accumulated in metaphase with colcemid alone or after treatment with hydroxyurea and Nocodazole. Under the experimental conditions used in this study, the latter procedure gives much better results, avoiding in particular the extensive formation of micronucleated cells. Aphidicolin, an inhibitor of DNA polymerase, in combination with Nocodazole, provides a useful method to tightly synchronize these cells at the G1/S border.

Characterization of subcellular components in synchronized hepatoma cells as a function of the cell cycle

The specific activity and subcellular distribution of marker enzymes for the main subcellular components were analysed in homogenates of synchronized hepatoma cells (Morris 7288c), obtained by selective detachment at mitosis combined with a metaphase block with Colcemid. Markers for lysosomes, mitochondrial outer membrane, plasma membrane and cytosol are synthesized throughout the cycle at the same rate as the bulk of cellular protein. Larger variations are observed for a Golgi marker; after a decrease around mitosis, the specific activity of galactosyltransferase increases steadily from middle G(1)-phase on, and at the end of G(2)-phase it is nearly twice that observed at the beginning of G(1)-phase. Our results show that synthesis of cytochrome oxidase may occur preferentially in G(2)-phase. Large modifications of the density distribution of lysosomes are observed during the cell cycle; the median equilibrium density of lysosomal markers decreases in G(1)-phase, and some increase in soluble activity occurs at the same time. Reverse changes occur progressively during S- and G(2)-phases. At mitosis, Golgi galactosyltransferase shows a more dispersed distribution, and modifications in the density distribution of endoplasmic-reticulum NADPH-cytochrome c reductase are observed. The latter can be most easily explained by a detachment of ribosomes from endoplasmic-reticulum membranes. No significant modifications occur in mitochondrial and plasma-membrane markers.

Glucocorticoid induction of tyrosine aminotransferase in cultured cells

For over a decade, tyrosine aminotransferase induction in tissue culture cells has been a useful model system in which to study glucocorticosteroid action. In the 1960s, the establishment in culture of rat hepatomas expressing the inducible enzyme, already known to be induced in liver in vivo, provoked a wide-ranging series of experiments. The data from these experiments have provided considerable information regarding the mechanism of action of steroids. These include the fundamental facts that the steroids act directly on the induced cell in unmetablized form, that removal of steroid results in deinduction, that induction does not require DNA synthesis or massive changes in RNA synthesis, and that cytoplasmic receptor occupancy by active steroids correlates closely with the steroids' ability to affect inductions. Studies in tissue culture cells have led to the analysis of transcriptional and posttranscriptional models attempting to explain enzyme induction. The effects on enzyme induction of nonsteroid hormones and other factors have been studied through the use of tissue culture cells. Finally, cells and clones of cell variants are being used to study enzyme induction, through biochemical analysis and cell genetic approaches, including somatic cell hybridization.

Posttranscriptional regulation of glucocorticoid-regulated functions

Relying heavily on studies of TAT regulation in cultured rat hepatoma cell lines, we have attempted in this brief review to discuss possible mechanisms for posttranscriptional regulation of glucocorticoid-sensitive enzymes and to chronicle the evidence for and against posttranscriptional mechanisms for specific enzyme induction by glucocorticoids. Initially, mechanisms were considered that would reconcile results showing sensitivity of both induction and deinduction of TAT to inhibitors of RNA synthesis with studies demonstrating first that glucocorticoids regulate the rates of specific enzyme synthesis and, then, that glucocorticoids regulate levels of enzyme-specific mRNA. Such reconciliation proved unnecessary when it was demonstrated that inhibitors of RNA synthesis such as actinomycin D were not specific for RNA synthesis, but also had effects on mRNA turnover and protein metabolism. The bulk of evidence to date establishes that glucocorticoids promote the production of enzyme-specific mRNA for the proteins whose synthesis is regulated by thses steroids. Nevertheless, there is still very little direct evidence that steroids can modulate rates of specific gene transcription. The glucocorticoid stimulation of mouse mammary tumor virus RNA production in cultured cell lines is the only example to date where such a mechanism is supported by RNA-DNA hybridization studies. Posttranscriptional actions of steroids on the turnover, processing, or extranuclear transport of specific mRNA precursors remain potential steps at which glucocorticoids might function. The rapid turnover of some glucocorticoid-regulated enzymes and their mRNAs not only ensures a rapid response to steroid addition or withdrawal, but also subjects these proteins to relatively large fluctuations upon alterations in overall protein or mRNA metabolism. Thus many of the inductions and repressions of hepatic TAT and TO by mediators other than the glucocorticoids may be attributable entirely to nonspecific mechanisms.

Some paradoxical effects of inhibitors of protein synthesis on protein turnover in cultured human cells

Low concentrations of cycloheximide, sufficient to block net protein synthesis in growing normal and cancer cells, had no effect on protein turnover, i.e. either the incorporation of labeled amino acids from media lacking other amino acids essential for growth, or the loss to the medium of amino acids from prelabeled cells. At the concentrations that blocked growth, the rate of amino acid incorporation from complete medium was reduced to the "turnover level" i.e. the rate of incorporation seen in amino acid-deficient media. Protein turnover was inhibited only at higher concentrations of the inhibitor. Qualitatively similar results have been obtained with puromycin, anisomycin, emetin and tylocerebrine.

The effects of metabolic inhibitors on the synthesis of inducible tyrosine aminotransferase in cultured hepatoma cells

The effects of actinomycin-D and 3'-deoxyadenosine (cordycepin) on the steroid-mediated induction of tyrosine aminotransferase (TAT) synthesis have been reexamined in view of recent reports that the primary inhibitory action of these compounds may affect synthesis of proteins as well as RNA. The present results confirm that cordycepin blocks the steroid-mediated induction of TAT in rat hepatoma cells (HTC), but unlike actinomycin-D, cordycepin neither increases nor maintains the levels of TAT found in HTC cells preinduced with dexamethasone. Indeed, cordycepin added to preinduced cells, either in the presence or absence of steroid, causes a prompt decline in TAT activity. These data also confirm that both actinomycin-D and cordycepin have an early inhibitory effect on protein synthesis, but the cordycepin effect is observed sooner and the extent of inhibition is greater. When actinomycin-D and cordycepin are added simultaneously to preinduced cells with the steroid removed, the actinomycin-td produced maintenance of preinduced levels of TAT persists. Also, the inhibition of protein synthesis in cultures with both inhibitors approaches that for the cells treated with actinomycin-D alone instead of cordycepin alone. These data suggest that cordycepin inhibits TAT synthesis in preinduced cells by its inhibition of protein synthesis, and this inhibitory effect of cordycepin is blocked by actinomycin-D. It is possible that actinomycin-D does this by preventing the incorporation of cordycepin into RNA. However, regardless of the correctness of this speculation, the multiple effects of cordycepin indicate that this inhibitor cannot be used either to prove or rule out the post-transcriptional model for regulation of gene expression. Also, this requirement that protein synthesis must continue in order to maintain pre-induced levels of TAT is inconsistent with the assumption that the maintenance of these induced TAT levels by actinomycin-D is due to inhibition of TAT degradation.

Analysis of the induction and deinduction of tyrosine aminotransferase in enucleated HTC cells

Anucleate HTC cells have been used to determine the importance of the nucleus in the regulation of the intracellular levels of tyrosine aminotransferase (TAT) in hepatoma tissue culture (HTC) cells. In the absence of the nucleus, neither the induction of the enzyme by dexamethasone nor its deinduction upon removal of the hormone occurs. Degradation of the enzyme takes place when protein synthesis is inhibited in anucleates by cycloheximide. Therefore, the maintenance of induced levels of enzyme activity after dexamethasone withdrawal from pre-induced anucleates suggest that the nucleus is required for the inactivation of the TAT mRNA.

Regulation of tyrosine aminotrasferase activity in two liver-derived permanent cell lines

The regulation of tyrosine aminotransferase (TAT) activity has been examined in two liver-derived heteroploid cell lines. One (hepatoma tissue culture cells [HTC]) was derived from a hepatoma, the other (rat liver culture cells [RLC]) was derived from normal liver. The two cell lines show the following striking similarities in the control of this specific protein: (a) The kinetics of TAT induction by dexamethasone phosphate (DxP) are similar in randomly growing cells of both lines; (b) During mitosis and early G(1) phase of the cell cycle TAT activity cannot be induced by DxP in either cell line; (c) 2-3 h into G(1), when both lines become sensitive to inducer, basal enzyme activity declines to a new steady-state level; (d) Preinduced cells collected in mitosis show approximately twice the level of TAT activity as fully induced, randomly growing cultures and this activity is maintained in early G(1) with or without the inducer; and (e) Inhibition of RNA synthesis by 5 microg/ml of actinomycin D in preinduced, synchronized cells allows TAT activity to remain at constitutive levels throughout G(1), even in the absence of inducer. These results are presented in support of a previously described model which states that glucocorticoid hormones exert posttranscriptional control of the synthesis of specific proteins in mammalian cells.

Suppression of DNA synthesis in hepatoma cells exposed to glucocorticoid hormone in vitro

Glucocorticoid hormone is shown to markedly suppress DNA synthesis in a line of rat hepatoma cells in vitro. In the presence of 300 nM hydrocortisone or 30 nM dexamethasone the incorporation of radioactive thymidine falls to 50% of control levels by 36 hr, and at higher concentrations of hormone inhibition can be noted as early as 12 hr and is nearly complete by 24 hr. This inhibition of radioactive thymidine incorporation reflects a true suppression of DNA synthesis, is accompanied by a corresponding inhibition of cell proliferation, and is readily reversible upon subsequent removal of hormone. In contrast to previously described effects of the glucocorticoid hormones on various cells of lymphoid origin, the inhibition of DNA synthesis in these hepatoma cells is not accompanied by appreciable cell lysis or by degradation of preformed DNA, and even when [(3)H]thymidine incorporation into DNA is inhibited by 90% or more, incorporation of [(14)C]uridine into RNA proceeds with little change. These findings all parallel previous observations on the effects of glucocorticoid hormone on the livers of intact animals and suggest that studies on the mechanism of the inhibition of DNA synthesis in the present more isolated system may lead to a better understanding of the means by which these compounds inhibit liver growth in vivo. Despite the ready suppressibility of DNA synthesis in these hepatoma cells and in two other cell lines of liver origin, none of these cell lines was found to be inducible for tyrosine aminotransferase. The apparent dissociation between two "steroid-sensitive" phenomena is of interest and warrants further investigation.

Expression of tyrosine aminotransferase activity in somatic-cell heterokaryons: evidence for negative control of enzyme expression

Cells of the tyrosine aminotransferase (EC 2.6.1.5)-inducible line, HTC, were fused with those of the BRL-62 line, which contain little or no enzyme and are unresponsive to the usual steroid inducers. Histochemical examination of the resulting heterokaryons within 24 hr of the fusion showed them to have little or no enzyme and to be unresponsive to steroids. These experiments disclose that the observed negative dominance is universal in the fused cells and that such dominance occurs promptly after fusion, when loss of chromosomes has not occurred.

Actinomycin-D enhancement of glutamine synthetase activity in chick embryo retinas cultured in the presence of cortisol

Cortisol can prematurely induce glutamine synthetase activity in the chick embryo retina. Under appropriate conditions, this effect can be enhanced by addition of low levels of actinomycin D; this enhancement is reversibly inhibited by cycloheximide. The magnitude of the effect is a function of time of exposure to hormone as well as antibiotic and is also a function of the age of the embryo; within the limits of the present study it did not appear to be a function of actinomycin-D concentration. The data are discussed in terms of current ideas of possible control mechanisms in animal cells.

Tyrosine transaminase induction by dexamethasone in a new rat liver cell line

A cell line derived from normal, adult rat liver has been established; the cells are similar to hepatocytes, as shown by electron microscopy. The addition of dexamethasone to the culture medium induced a three- to sixfold increase in the specific activity of tyrosine alpha-ketoglutarate transaminase; this increase was inhibited by the simultaneous addition of cycloheximide or actinomycin D. The latter, when added to cells given prior treatment with dexamethasone, further enhanced the transaminase activity. Contact-inhibited cells showed a lower response to dexamethasone than exponentially growing cells.

Glucocorticoid-induced alteration of the surface membrane of cultured hepatoma cells

Glucocorticoids induce an alteration of the surface of hepatoma tissue culture (HTC) cells as expressed by changes in cell electrophoretic, antigenic, and adhesive properties. The alteration is assayed by the increased adhesiveness of induced cells for a glass surface. The induction process has a lag period of about 3 hr and attains a plateau level after 24-30 hr when 50-80% of the steroid-treated cells are firmly adhered. Less than 10% of untreated cells adhere under the same conditions. Induction is inhibited by actinomycin D and cycloheximide, demonstrates both pH and temperature dependence, and responds to changes in steroid concentration and structure. By contrast, the attachment per se of preinduced cells is not affected by inhibitors of RNA and protein synthesis, fluctuations of temperature and pH, and the presence or absence of the hormone. When the induction process is reversed by removal of steroid or addition of actinomycin D, preinduced adhesiveness is lost with a half-life of 13-24 hr, but in the presence of cycloheximide the loss is accelerated (t(1/2) 3-5.5 hr). These results suggest that glucocorticoids induce the biosynthesis of a protein which either modifies the cell surface (an enzyme) or is incorporated into surface structures (structural protein).

Posttranscriptional control of tyrosine aminotransferase synthesis by insulin

Adrenal steroid hormones induce the synthesis of tyrosine amino-transferase (TAT) in HTC cells, an established line of rat hepatoma cells in tissue culture. The addition of insulin to cells previously induced in a serum-free medium results in a rapid two- to threefold further increase in the rate of synthesis of TAT and a small increase in total amino acid incorporation. These changes do not require concomitant RNA synthesis, suggesting that insulin acts at a step in protein synthesis beyond that of gene transcription. Although the effects of insulin on HTC cells are similar to those caused by dialyzed bovine serum, evidence is presented that insulin and serum affect different aspects of TAT synthesis. Unlike the glucocorticoids, insulin does not cause the accumulation of TAT mRNA, nor the sustained induction of TAT. The continued presence of the inducing steroid is required to permit maximal expression of the insulin effect.

The appearance and disappearance of the post-transcriptional repressor of tyrosine aminotransferase synthesis during the HTC cell cycle

The synthesis of the enzyme tyrosine aminotransferase in HTC cells (an established line of rat hepatoma cells) is inducible by glucocorticoid hormones only during the latter part of G1 phase and throughout S phase in the cell generation cycle. We have earlier shown that during the first few hours of G1 phase when the enzyme cannot be induced, its synthesis is constitutive, presumably using as template, preexisting messenger RNA. Our model for tyrosine aminotransferase gene regulation in eukaryotic cells entails a specific post-transcriptional repressor which is formed only during the periods in the cell cycle when tyrosine aminotransferase is inducible. This model predicts that during the noninducible period, G2, the tyrosine aminotransferase repressor would not be present and thus tyrosine aminotransferase synthesis would be constitutive. Data are presented which confirm this prediction in further support of the model.

Control of tyrosine aminotransferase synthesis in tissue culture by a factor in serum

The hormonal induction of tyrosine aminotransferase in an established line of rat hepatoma cells is enhanced by the presence of a macromolecular component of serum in the inducing medium. The addition of serum to cells previously induced in serum-free medium results in a rapid two- to threefold increase in the rate of tyrosine aminotransferase synthesis, as measured by specific radio-immunoprecipitation techniques, as well as a smaller increase in over-all protein synthesis. The stimulation of protein synthesis by serum is accompanied by an increase in the proportion of ribosomes sedimenting as polysomal aggregates. The increase in the rate of TAT synthesis is largely independent of RNA synthesis as it is insensitive to antinomycin D, suggesting that the serum acts at a site beyond gene transcription. The maximum effect of serum on TAT synthesis requires the continued presence of the hormone inducer.