Interaction of Glucocorticoid Hormones and Cyclic Nucleotides in Induction of Tyrosine Aminotransferase in Cultured Hepatoma Cells *

Reproducible induction of the enzyme tyrosine aminotransferase by dibutyryl CAMP (Bt,cAMP) in a line of HTC hepatoma cells in suspension culture requires that the cells be preinduced with dexamethasone, a synthetic glucocorticoid which itself induces tyrosine aminotransferase. Concentrations of dexamethasone that do not induce tyrosine aminotransferase fail to support B&CAMP induction, removal of the steroid from the medium leads to a loss of the Bt,cAMP effect, and an HTC cell line whose aminotransferase is not steroid-inducible does not respond to the cyclic nucleotide. We show that the further induction of tyrosine aminotransferase by Bt,cAMP in dexamethasone-treated cells is due to an increased rate of enzyme synthesis. The cyclic nucleotide has no effect on aminotransferase synthesis in cells grown in the absence of steroid. Several lines of evidence suggest that dexamethasone acts at a step beyond the activation of protein kinase by CAMP: (a) basal levels of CAMP are not altered by growth of HTC cells in dexamethasone; (6) accumulation of CAMP from the medium is not enhanced; (c) the glucocorticoid does not induce CAMP-dependent protein kinase in HTC cells; and (d) there is no augmentation of CAMP binding to the regulatory protein, nor is there any change in CAMP activation of protein kinase caused by growth in dexamethasone. These results help define a system that should be useful in studying the interaction of cyclic nucleotides and steroid hormones.


Reproducible
induction of the enzyme tyrosine aminotransferase by dibutyryl CAMP (Bt,cAMP) in a line of HTC hepatoma cells in suspension culture requires that the cells be preinduced with dexamethasone, a synthetic glucocorticoid which itself induces tyrosine aminotransferase.
Concentrations of dexamethasone that do not induce tyrosine aminotransferase fail to support B&CAMP induction, removal of the steroid from the medium leads to a loss of the Bt,cAMP effect, and an HTC cell line whose aminotransferase is not steroid-inducible does not respond to the cyclic nucleotide. We show that the further induction of tyrosine aminotransferase by Bt,cAMP in dexamethasone-treated cells is due to an increased rate of enzyme synthesis. The cyclic nucleotide has no effect on aminotransferase synthesis in cells grown in the absence of steroid.
Several lines of evidence suggest that dexamethasone acts at a step beyond the activation of protein kinase by CAMP: (a) basal levels of CAMP are not altered by growth of HTC cells in dexamethasone; (6) accumulation of CAMP from the medium is not enhanced; (c) the glucocorticoid does not induce CAMP-dependent protein kinase in HTC cells; and (d) there is no augmentation of CAMP binding to the regulatory protein, nor is there any change in CAMP activation of protein kinase caused by growth in dexamethasone.
These results help define a system that should be useful in studying the interaction of cyclic nucleotides and steroid hormones.
The concept that adrenal glucocorticoid hormones allow key metabolic processes to achieve maximal rates was first proposed by Ingle (1). This role of glucocorticoids has subsequently been shown to be important in lipolysis (2,3), gluconeogenesis (4, 51, glycogenesis (61, and glycogenolysis (7). These processes are primarily regulated by peptide and/or catecholamine hormones, presumably through the action of CAMP. Although steroid hormones are thought to act by mechanisms *  separate from the CAMP-dependent protein kinase system (81, the examples cited above indicate a definite interplay of glucocorticoids and cyclic nucleotides, although the biochemical site of this interaction remains obscure. Induction by glucocorticoids of the hepatic enzyme tyrosine aminotransferase (L-tyrosine:2-oxoglutarate aminotransferase, EC 2.6.1.5), which is rate-limiting for tyrosine degradation, has served as a valuable model for studying the gluconeogenie action of this class of hormones (9). In addition, CAMP or its dibutyryl derivative induces this enzyme in intact adult liver (lo), fetal liver in organ culture (ll), and in some hepaticderived tissue culture lines (12)(13)(14). Although an early report suggested that glucocorticoid induction of tyrosine aminotransferase is independent of CAMP (151, the converse is not true. Several studies suggest that CAMP-mediated induction of this enzyme is not maximal unless glucocorticoids are also present. For example, adrenalectomy significantly blunts the inductive effect of CAMP or analogs on rat liver tyrosine aminotransferase (10,16,17). Also, combinations of glucocorticoids and dibutyryl CAMP result in a synergistic aminotransferase induction in rat liver (10) and fetal liver organ culture (11). Finally, cortisol has been shown to potentiate the induction of tyrosine aminotransferase by B&AMP' in the cultured H35 hepatoma cell system (12). This requirement for glucocorticoids for maximal CAMP-mediated aminotransferase induction has recently been clarified in the well-defined HTC cultured hepatoma cell system. In suspension cultures of HTC cells, significant induction of tyrosine aminotransferase cannot be routinely achieved (12,14,15), whereas in monolayer cultures a small effect is observed (13,18). Restoration of the full responsiveness of HTC cells grown in suspension culture to' butyrylated derivatives of CAMP can be accomplished by growing the cells in culture medium containing a glucocorticoid hormone for a period of time prior to addition of the cyclic nucleotide (19). In this paper we provide a detailed description of this phenomenon, which should prove useful in studying the interaction of steroid hormones and cyclic nuclebtides.

EXPERIMENTAL PROCEDURES
Cell Culture-HTC cells were grown in suspension culture in a modified Swim's S77 (S77S) medium as described previously (14). is one of several isolated, in the absence of selective pressure, by serial subclonings of the wild type HTC cell line (20). These cells possess basal levels of aminotransferase which can be precipitated or neutralized by anti-aminotransferase antibody.
They also possess glucocorticoid receptors which, by several criteria, appear to be the same as those of wild type cells, but the aminotransferase of these cells is virtually not induced by glucocorticoids (20). Enzyme Asxzys-The activity of tyrosine aminotransferase was determined by a modification of the method of Diamondstone (21) and protein concentration was determined by the method of Lowry et al. (22). Results are expressed as specific activity (E,;,) in terms of milliunits (nanomoles of p-hydroxphenylpyruvate) formedimg of proteinimin.
The preparation of cytosol extracts of HTC cells and assays for protein kinase and CAMP binding activities were performed as described in detail previously (23 There is, however, no change in the affinity of binding of ["HIcAMP to the regulatory protein, nor is there any increase in the maximal binding capacity of this receptor (Fig.  5A), as has been suggested to happen in prolactin action on mammary gland explants (34). The extent of CAMP-mediated stimulation of protein kinase is unaltered by growth in S77S + dexamethasone, and the concentration of CAMP required for half-maximal activation is identical in both types of cells (Fig.  .5B). There thus would appear to be no glucocorticoid-mediated alteration in the interaction of the regulatory and catalytic subunits of protein kinase. Although none of the techniques employed would be expected to detect a small fraction of the total protein kinase activity that might be specifically involved in tyrosine aminotransferase induction, there is no evidence to suggest that such a protein exists. We therefore conclude from these experiments that the ability of steroid-treated HTC cells to respond to Bt,cAMP is not due to the restoration of a liver-type protein kinase. HTC cells were grown in S77S i dexamethasone for 4 days. The procedures used to prepare cytosol extracts from these cells and liver, and for the chromatography were described previously (14). Protein kinase assays were performed on aliquots from each fraction in the presence of 1 FLM CAMP. The symbols indicate protein kinase activity from liver, HTC cells grown in S77S, and in S77S + 5 PM dexamethasone.

DISCUSSION
The important role products of the adrenal cortex play in regulating carbohydrate metabolism (35,36) was well established by the time the so-called glucocorticoid hormones were specifically identified in the late 1940's. Shortly thereafter it was recognized that such hormones are involved in a variety of seemingly unrelated processes, leading Ingle to postulate that the glucocorticoids are general hormones which allow tissues to sustain homeostasis by permitting maximal rates of critical metabolic processes (1). The classical examples of this requirement for glucocorticoids are gluconeogenesis, glycogenolysis, and glycogenesis, processes on which the steroid has little or no effect (4-7). Steroids themselves affect lipolysis and inhibition of glucose entry into cells and in addition greatly magnify the effect of other hormones on these processes (37-39). The hormones which primarily regulate these processes, epinephrine, glucagon, and growth hormone, all are thought to act by increasing intracellular CAMP concentrations and/or activity of the nucleotide. In addition other CAMP-mediated hormone effects, such as the action of catecholamines on the cardiovascular system (40) and on the reaction of skin to cold (41,42) as well as amino acid transport in liver cells (43), appear to depend on glucocorticoids.
The mechanism of the apparent interaction of the steroid and peptide hormones (presumably acting through CAMP) has not been explained in any of these cases (44).
The experiments which suggest that glucocorticoids and cyclic nucleotides interact in regulating tyrosine aminotransferase activity were cited in the introduction. The availability of a tissue culture cell line in which consistently reproducible CAMP effects require that steroid be present allows for a detailed analysis of this interplay.
One can approach this problem either from the manner by which CAMP could influ-  16,46), as has the observation that the effect of steroids occurs at a step beyond CAMP (4, 7). Further support for this notion is given by the observation that when HTC cells are made sensitive to cyclic nucleotides, the effective concentration range for these compounds is the same as in RLC cells, a line that is normally responsive (14). Since CAMP is thought to act in eukaryotic cells $y stimulating protein kinases, an effect of the glucocorticoid on this complex seemed possible, particularly since HTC cells lack the major CAMP-dependent protein kinase of liver (23,33 Glucocorticoid-Cyclic AMP Interaction 3897