Thyroid hormone induction of an autocrine growth factor secreted by pituitary tumor cells.

Thyroid hormones stimulate the rate of cell division by poorly understood mechanisms. The possibility that thyroid hormones increase cell growth by stimulating secretion of a growth factor was investigated. Thyroid hormones are nearly an absolute requirement for the division of GH4C1 rat pituitary tumor cells plated at low density. Conditioned media from cells grown with or without L-triiodothyronine (T3) were treated with an ion exchange resin to remove T3 and were tested for ability to stimulate the division of GH4C1 cells. Conditioned medium from T3-treated cells was as active as thyroid hormone at promoting GH4C1 cell growth but did not elicit other thyroid hormone responses, induction of growth hormone, and down-regulation of thyrotropin-releasing hormone receptors, as effectively as T3 did. A substance or substances associated with T3-induced growth stimulatory activity migrated at high molecular weight at neutral pH and was different from known growth-promoting hormones induced by T3. The results demonstrate that thyroid hormones stimulate the division of GH4C1 pituitary cells by stimulating the secretion of an autocrine growth factor.

Solubilization of pituitary receptors for thyrotropin-releasing hormone.

Receptors for thyrotropin-releasing hormone were solubilized by Triton X-100. Membrane fractions from GH3 pituitary tumor cells were incubated with thyrotropin-releasing hormone in order to saturate specific receptor sites before the addition of detergent. The amount of protein-bound hormone solubilized by Triton X-100 was proportional to the fractional saturation of specific membrane receptors. Increasing detergent:protein ratios from 0.5 to 20 led to a progressive loss of hormone . receptor complex from membrane fractions with a concomitant increase in soluble protein-bound hormone. The soluble hormone . receptor complex was not retained by 0.22 micron filters and remained soluble after ultracentrifugation. Following incubation with high (2.5--10%) concentrations of Triton X-100 and other non-ionic detergents, or following repeated detergent extraction, at least 18% of specifically bound thyrotropin-releasing hormone remained associated with particulate material. Unlike the hormone receptor complex, the free hormone receptor was inactivated by Triton X-100. A 50% loss of binding activity was obtained with 0.01% Triton X-100, corresponding to a detergent:protein ratio of 0.033. The hormone . receptor complex was included in Sepharose 6B and exhibited an apparent Stoke radius of 46 A in buffers containing Triton X-100. The complex aggregated in detergent-free buffers. Soluble hormone receptors were separated from excess detergent and thyrotropin-releasing hormone by chromatography on DEAE-cellulose. Thyrotropin-releasing hormone dissociated from soluble receptors with a half-time of 120 min at 0 degrees C, while the membrane hormone . receptor complex was stable for up to 5 at 0 degrees C.

Characterization of the calcium response to thyrotropin-releasing hormone in lactotrophs and GH cells.

Thyrotropin-releasing hormone (TRH) acts via a G-protein-coupled receptor on lactotrophs to increase the intracellular free calcium ion concentration, [Ca(2+)](i). The [Ca(2+)](i) response depends on both TRH concentration and the duration of TRH exposure. An initial, short-lived [Ca(2+)](i) spike results from release of Ca(2+) from intracellular stores, whereas a later sustained [Ca(2+)](i) increase, often characterized by [Ca(2+)](i) oscillations, results from an influx of extracellular Ca(2+) through both voltage-gated and non-voltage-gated, store-operated Ca(2+) channels. The initial spike phase predominates at high doses of TRH, whereas the plateau phase predominates at low doses. The mechanisms underlying the complex [Ca(2+)](i) response to TRH are discussed.

Pituitary thyrotropin-releasing hormone (TRH) receptors: effects of TRH, drugs mimicking TRH action, and chlordiazepoxide.

Binding of TRH to specific cell surface receptors on clonal GH4C1 cells is followed within 10 min by receptor sequestration and over 24 h by receptor down-regulation. These experiments were designed to determine if TRH-activated second messenger systems are responsible for changes in receptor localization or number. BAY K8644 and A23187, which increase intracellular calcium, alone or together with 12-O-tetradecanoyl phorbol acetate (TPA), which activates protein kinase C, did not appear to internalize TRH receptors. Drug treatment did not alter the rate of [3H]MeTRH association or internalization, determined by resistance to an acid/salt wash, or the amount of [3H]MeTRH able to bind at 0 C, where only surface receptors are accessible. TPA (0-100 nM) alone or in combination with BAY K8644 or A23187, also failed to change receptor number or affinity after 48 h when TRH caused a 75% decrease in the density of specific binding sites. Chlordiazepoxide has been reported antagonize TRH binding and TRH-induced phospholipid breakdown. Chlordiazepoxide shifted the dose-response curves for TRH stimulation of PRL release and synthesis to the right, and did not change PRL release alone. The affinity of receptors for chlordiazepoxide was not affected by a nonhydrolyzable analog of GTP whereas affinity for TRH was decreased; these properties are consistent with the classification of chlordiazepoxide as a competitive antagonist. Several experiments tested whether chlordiazepoxide would cause receptor internalization and down-regulation. Chlordiazepoxide did not appear to internalize TRH receptors, because TRH-binding sites became available rapidly and at the same rate after they had been saturated with chlordiazepoxide at 0 or 37 C.(ABSTRACT TRUNCATED AT 250 WORDS)

Signal transduction, desensitization, and recovery of responses to thyrotropin-releasing hormone after inhibition of receptor internalization.

Three independent methods were used to block internalization of the TRH receptor: cells were infected with vaccinia virus encoding a dominant negative dynamin, incubated in hypertonic sucrose, or stably transfected with a receptor lacking the C-terminal tail. Internalization was blocked in all three paradigms as judged by microscopy using a fluorescently labeled TRH agonist and biochemically. The initial inositol trisphosphate (IP3) and Ca2+ responses to TRH were normal when internalization was inhibited. The IP3 increase was sustained rather than transient, however, in cells expressing the truncated TRH receptor, implying that the C-terminal tail of the receptor may be important for uncoupling from phospholipase C. After withdrawal of TRH, cells were refractory to TRH until both ligand dissociation and resensitization of the receptor had occurred. When surface-bound TRH was removed by a mild acid wash, which did not impair receptor function, neither wild-type nor truncated receptors were able to generate full IP3 responses for about 10 min. The rate of recovery was not altered by blocking internalization. Recovery of intracellular Ca2+ responses also depended on the rate of Ca2+ pool refilling. In summary, in the continued presence of TRH, phospholipase C activity declines quickly due to receptor uncoupling; this desensitization does not take place for the truncated receptor. After TRH is withdrawn, cells are refractory to TRH. Before cells can respond, TRH must dissociate and a resensitization step, which takes place on the plasma membrane and does not require the C-terminal tail of the receptor, must occur.

Thyrotropin (TSH)-releasing hormone stimulates TSH beta promoter activity by two distinct mechanisms involving calcium influx through L type Ca2+ channels and protein kinase C.

TRH stimulates rat (r) TSH beta gene promoter activity at two distinct response elements, which also respond to protein kinase C-signaling pathways. The dependence of TRH-stimulated transcription of the TSH beta gene on a rise in intracellular calcium [Ca2+]i, and on the necessity for Ca2+ influx through L-type voltage-gated calcium channels was investigated in two transfected cell lines and in normal thyrotropes. The transcription rate of the homologous gene in normal thyrotropes was measured by nuclear run-off assays. Bay K8644, an L channel agonist, stimulated TSH beta gene transcription 6-fold, and TRH stimulation of TSH beta gene transcription was partially blocked by nimodipine, an L channel antagonist, while phorbol 12-myristate-13-acetate (PMA)-stimulated transcription was not. Bay K8644 plus TRH had a greater effect than either treatment alone. Constructs of the 5'-flanking region of the TSH beta gene fused to the luciferase reporter (TSH beta LUC) were then transfected into excitable GH3 pituitary cells. TSH beta LUC was stimulated 2- to 5-fold by 1 nM TRH or 100 nM Bay K8644, and the TRH effect was nearly abolished by nimodipine or chelation of external Ca2+. Constructs containing isolated TRH-responsive elements fused to a heterologous promoter responded similarly. The protein kinase C activator, PMA (100 nM) also stimulated TSH beta LUC transcription, but its effect was not inhibited by nimodipine. A stable heterologous cell line containing the mouse TRH receptor was constructed by transfection of nonexcitable 293 cells, which lack L channel activity. In the resultant 301 cells, TSH beta LUC activity was increased 2- to 3-fold by TRH or PMA; nimodipine, Bay K8644, and removal of extracellular Ca2+ had no effect. We conclude that TRH stimulation of TSH beta gene transcription requires Ca2+ release from inositol triphosphate-sensitive stores and Ca2+ influx via L-type calcium channels in GH3 cells, but in transfected 293 cells TRH activation of protein kinase C plays a predominant role in activating TSH beta. Both mechanisms appear to be operative in normal thyrotropes.

Activation of MAPK by TRH requires clathrin-dependent endocytosis and PKC but not receptor interaction with beta-arrestin or receptor endocytosis.

To determine whether the interaction of the TRH receptor with beta-arrestin is necessary for TRH activation of MAPK, cells expressing either intact or truncated, internalization-defective TRH receptors were transfected with a beta-arrestin-green fluorescent protein conjugate. In cells expressing the wild-type pituitary TRH receptor, TRH caused translocation of the beta-arrestin-green fluorescent protein conjugate from the cytosol to the plasma membrane within 30 sec. After 5 min, the beta-arrestin-green fluorescent protein conjugate was visible in vesicles, where it colocalized with rhodamine-labeled TRH. In hypertonic sucrose, the beta-arrestin-green fluorescent protein conjugate translocated to the plasma membrane after TRH addition but did not internalize. In cells expressing the truncated TRH receptor, TRH did not cause translocation of the beta-arrestin-green fluorescent protein conjugate. TRH activated MAPK strongly in cells expressing intact or truncated TRH receptors, indicating that the receptor does not need to bind beta-arrestin or internalize. MAPK activation by TRH, epidermal growth factor, and phorbol ester was strongly inhibited by hypertonic sucrose and concanavalin A, which block movement of proteins into coated pits and coated pit assembly. Hypertonic sucrose did not affect MAPK activation in cells overexpressing MAPK kinase 1. Dominant negative dynamin, which blocks conversion of coated pits to vesicles, also reduced receptor internalization and TRH activation of MAPK. TRH activation of MAPK required PKC but was insensitive to pertussis toxin and did not require ras, epidermal growth factor receptor kinase, or PI3K. These results show that the TRH receptor itself does not need to bind beta-arrestin or undergo sequestration to activate MAPK but that the endocytic pathway must be intact.

Expression cloning and characterization of PREB (prolactin regulatory element binding), a novel WD motif DNA-binding protein with a capacity to regulate prolactin promoter activity.

Previous studies have implied that a transcription factor(s) other than Pit-1 is involved in homeostatic regulation of PRL promoter activity via Pit-1-binding elements. One such element, 1P, was employed to clone from a rat pituitary cDNA expression library a novel 417-amino acid WD protein, designated PREB (PRL regulatory element binding) protein. PREB contains two PQ-rich potential transactivation domains, but no apparent DNA-binding motif, and exhibits sequence-specific binding to site 1P, to a site nonidentical to that for Pit-1. The PREB gene (or a related gene) is conserved, as an apparently single copy, in rat, human, fly, and yeast. A single approximately 1.9-kb PREB transcript accumulates in GH3 rat pituitary cells, to levels similar to Pit-1 mRNA. PREB transcripts were detected in all human tissues examined, but the observation of tissue-specific multiple transcript patterns suggests the possibility of tissue-specific alternative splicing. RT-PCR analysis of human brain tumor RNA samples suggested region-specific expression of PREB transcripts in brain. Western and immunocytochemical analysis implied that PREB accumulates specifically in GH3 cell nuclei. Transient transfection employing PREB-negative C6 rat glial cells showed that PREB is as active as, and additive with, Pit-1 in transactivation of a PRL promoter construct, and that PREB, but not Pit-1, can mediate transcriptional activation by protein kinase A (PKA). Expression in GH3 cells of a GAL4-PREB fusion protein both strongly transactivated a 5XGAL indicator construct and yielded a further stimulation of expression of this construct by coexpressed PKA, implying that PREB can mediate both basal and PKA-stimulated transcriptional responses in pituitary cells. These observations imply that PREB will prove to play a significant transcriptional regulatory role, both in the pituitary and in other organs in which transcripts of its gene are expressed.

Pituitary thyrotropin-releasing hormone receptors: local anesthetic effects on binding and responses.

Pharmacological agents are widely used to probe the mechanism of action of TRH. A number of these drugs behave as local anesthetics at high concentrations. The effect of local anesthetics on the binding of [3H]Me-TRH to specific receptors was studied using the GH4C1 line of rat pituitary tumor cells. [3H]Me-TRH binding was inhibited by classical local anesthetics with the order of potency (IC50 values): dibucaine (0.37 mM) greater than tetracaine (1.2 mM) greater than lidocaine (3.3 mM) greater than procaine and benzocaine (greater than 10 mM). IC50 values for other drugs with local anesthetic properties that inhibited [3H]Me-TRH were: 100 microM trifluoperazine, 100 microM imipramine, 170 microM chlorpromazine, 300 microM verapamil, and 700 microM propranolol. Inhibition by tetracaine and verapamil increased as the pH was raised from 6 to 8.5, indicating that the free base form of the amine drugs was the inhibitory species, and the local anesthetic effect was greater at 37 C than at 24 C or 0 C. [3H]Me-TRH binding to receptors in isolated membranes was inhibited to the same extent as binding to receptors on intact cells. Local anesthetics were 3- to 20-fold less potent at inhibiting [3H]Me-TRH to digitonin-solubilized receptors than binding to intact cells. In contrast, the potency of chlordiazepoxide, a putative TRH antagonist, to inhibit [3H]Me-TRH binding was equal using cells and solubilized receptors (IC50 = 10 microM). Local anesthetics inhibited TRH-stimulated PRL release and also inhibited basal PRL secretion and secretion stimulated by two nonhormonal secretagogues, (Bu)2cAMP and a phorbol ester.(ABSTRACT TRUNCATED AT 250 WORDS)

Characterization of the calcium response to thyrotropin-releasing hormone (TRH) in cells transfected with TRH receptor complementary DNA: importance of voltage-sensitive calcium channels.

TRH stimulates a biphasic increase in intracellular free calcium ion, [Ca2+]i. Cells stably transfected with TRH receptor cDNA were used to compare the response in lines with and without L type voltage-gated calcium channels. Rat pituitary GH-Y cells that do not normally express TRH receptors, rat glial C6 cells, and human epithelial Hela cells were transfected with mouse TRH receptor cDNA. All lines bound similar amounts of [3H][N3-Me-His2]TRH with identical affinities (dissociation constant = 1.5 nM). Both pituitary lines expressed L type voltage-gated calcium channels; depolarization with high K+ increased 45Ca2+ uptake 20- to 25-fold and [Ca2+]i 12- to 14-fold. C6 and Hela cells, in contrast, appeared to have no L channel activity. GH4C1 cells responded to TRH with a calcium spike (6-fold) followed by a sustained second phase. When TRH was added after 100 nM nimodipine, an L channel blocker, the initial calcium burst was unaffected but the second phase was abolished. GH-Y cells transfected with TRH receptor cDNA responded to TRH with a 6-fold [Ca2+]i spike followed by a plateau phase (>8 min) in which [Ca2+]i remained elevated or increased. Nimodipine did not alter the peak TRH response or resting [Ca2+]i but reduced the sustained phase, which was eliminated by chelation of extracellular Ca2+. In the transfected glial C6 and Hela cells without calcium channels, TRH evoked transient, monophasic 7- to 9-fold increases in [Ca2+]i, and [Ca2+]i returned to resting levels within 3 min. Thapsigargin stimulated a gradual, large increase in [Ca2+]i in transfected C6 cells, and subsequent addition of TRH caused no further rise. Removal of extracellular Ca2+ from transfected C6 cells shortened the [Ca2+]i responses to TRH, to endothelin 1, and to thapsigargin. The TRH responses were pertussis toxin-insensitive. In summary, TRH can generate a calcium spike in pituitary, C6, and Hela cells transfected with TRH receptor cDNA, but the plateau phase of the [Ca2+]i response is not observed when the receptor is expressed in a cell line without L channel activity.

Calcium responses to thyrotropin-releasing hormone, gonadotropin-releasing hormone and somatostatin in phospholipase css3 knockout mice.

These studies examined the importance of phospholipase Cbeta (PLCbeta) in the calcium responses of pituitary cells using PLCbeta3 knockout mice. Pituitary tissue from wild-type mice contained PLCbeta1 and PLCbeta3 but not PLCbeta2 or PLCbeta4. Both Galphaq/11 and Gbetagamma can activate PLCbeta3, whereas only Galphaq/11 activates PLCss1 effectively. In knockout mice, PLCbeta3 was absent, PLCbeta1 was not up-regulated, and PLCbeta2 and PLCbeta4 were not expressed. Since somatostatin inhibited influx of extracellular calcium in pituitary cells from wild-type and PLCbeta3 knockout mice, the somatostatin signal pathway was intact. However, somatostatin failed to increase intracellular calcium in pituitary cells from either wild-type or knockout mice under a variety of conditions, indicating that it did not stimulate PLCbeta3. In contrast, somatostatin increased intracellular calcium in aortic smooth muscle cells from wild-type mice, although it evoked no calcium response in cells from PLCbeta3 knockout animals These results show that somatostatin, like other Gi/Go-linked hormones, can stimulate a calcium transient by activating PLCbeta3 through Gbetagamma, but this response does not normally occur in pituitary cells. The densities of Gi and Go, as well as the relative concentrations of PLCbeta1 and PLCbeta3, were similar in cells that responded to somatostatin with an increase in calcium and pituitary cells. Calcium responses to 1 nM and 1 microM TRH and GnRH were identical in pituitary cells from wild-type and PLCbeta3 knockout mice, as were responses to other Gq-linked agonists. These results show that in pituitary cells, PLCbeta1 is sufficient to transmit signals from Gq-coupled hormones, whereas PLCbeta3 is required for the calcium-mobilizing actions of somatostatin observed in smooth muscle cells.

Calcium channel agonists and antagonists: effects of chronic treatment on pituitary prolactin synthesis and intracellular calcium.

PRL synthesis by GH cells in culture has previously been shown to increase when calcium is added to cultures grown in calcium-depleted medium or when cultures are treated for 18 h or longer with the dihydropyridine calcium channel agonist BAY K8644, whereas the antagonist nimodipine inhibits PRL. The experiments described here were designed to test whether differences in PRL synthesis caused by the dihydropyridines are due to changes in PRL mRNA levels, whether structurally different classes of calcium channel blockers alter PRL production, and whether long term treatment with calcium channel agonists and antagonists alters intracellular free calcium, [Ca2+]i. PRL synthesis and PRL mRNA levels were increased similarly by BAY K8644 and decreased in parallel by the dihydropyridine antagonist nimodipine, while overall protein and RNA synthesis were not changed by either the agonist or antagonist. Two calcium channel blockers which act at different sites on L-type channels than the dihydropyridines also inhibited PRL synthesis without affecting GH; 5 microM verapamil reduced PRL by 64% and 15 microM diltiazem by 89%. Partial depolarization with 5-25 mM KCl increased PRL synthesis up to 2-fold. The intracellular free calcium ion concentration was estimated by Quin 2 and averaged 142 nM for control cultures in normal medium, and 128 and 168 nM for cultures treated 72 h with nimodipine or BAY K8644, respectively. Nimodipine totally prevented the calcium rise obtained upon depolarization.(ABSTRACT TRUNCATED AT 250 WORDS)

Attenuation of thyroid hormone action by 1,25-dihydroxyvitamin D3 in pituitary cells.

Interactions between thyroid hormone and 1,25-dihydroxyvitamin D3 [1,25-(OH)2D3] were examined in a rat pituitary tumor cell line, GH4C1. Cells were incubated in thyroid hormone-depleted medium for 2 days, and specific nuclear binding of [125I]T3 was measured. 1,25-(OH)2D3 decreased nuclear [125I]T3 binding without changing total cellular uptake of [125I]T3. This 1,25-(OH)2D3 effect required 2-3 h to become evident and 24 h to reach a maximum (40-50% of control) and was reversible. Treatment with 1,25-(OH)2D3 for 8 h changed the maximal binding capacity for [125I]T3 from 80.2 +/- 2.9 to 50.3 +/- 6.3 fmol/10(6) cells, whereas Kd was not significantly altered. The decrease in [125I]T3 binding was dose dependent, with an IC50 for 1,25-(OH)2D3 of 1 nM in thyroid hormone-depleted medium. 1,25-(OH)2D3 caused little change in [125I]T3 binding to isolated nuclei, i.e. 1,25-(OH)2D3 does not compete directly with [125I]T3 for binding. It is unlikely that 1,25-(OH)2D3 decreased [125I]T3 binding by increasing the concentration of intracellular free calcium ([Ca2+]i), since 1,25-(OH)2D3 did not change [Ca2+]i in Indo-I-loaded GH4C1 cells. Two major species (6 and 2.6 kilobases) of mRNA for c-erb-A, which have been reported to encode nuclear thyroid hormone receptors, were found by Northern blot analysis, and both were decreased by treatment with 1,25-(OH)2D3 for 8 h. T3 (2 nM) caused a 3-fold increase in GH production over 72 h and 1,25-(OH)2D3 inhibited GH induction by T3, with an IC50 at approximately 1 nM. 1,25-(OH)2D3 stimulated PRL synthesis 5-fold when 10 nM T3 was present, but not when T3 was absent. In summary, 1,25-(OH)2D3 caused a dose-dependent down-regulation of nuclear thyroid hormone receptors at a pretranslational level and diminished GH induction by T3. These results suggest that 1,25-(OH)2D3 inhibits GH synthesis indirectly, at least partly, by attenuating endogenous thyroid hormone action.

Differential effects of thyrotropin-releasing hormone, vasoactive intestinal peptide, phorbol ester, and depolarization in GH4C1 rat pituitary cells.

The sequence of PRL and GH release from GH4C1 cells was studied in perifusion and static culture systems. The secretory pattern elicited by TRH differed from those caused by depolarizing concentrations of KCl (Ca2+-initiated secretion), vasoactive intestinal peptide (VIP), 8-bromo-cAMP, and forskolin (cAMP-mediated secretion), and 12-O-tetradecanoylphorbol-13-acetate (TPA) (protein kinase C activation). TRH, K+, VIP, and TPA all caused secretion within 1 min in the perifusion system but the peak response to TRH and depolarization occurred earlier than the peak responses to TPA and VIP. PRL and GH release in response to a pulsatile application of TRH (0.4-min pulse every 5 min for 25 min) did not decline with a low dose, indicating that acute desensitization does not occur, but did decrease with a high concentration. When cells in the perifusion system were subjected to continuous stimulation, TRH caused a biphasic response with a 2- to 3-min period of high secretion followed by a second phase in which GH and PRL secretion were 60-70% the rates in the first phase. KCl caused predominantly first-phase secretion, and TPA caused a biphasic secretory pattern with a delay in its peak of action. VIP caused a modest but prolonged response whether administered in a pulsatile or sustained manner. When GH-cells were exposed to 100 nM TRH for 2 days, [3H] [N3-methyl-His2]TRH binding was decreased (down-regulation), intracellular PRL was increased (170% of control), and intracellular GH was decreased (65% of control). In these down-regulated cells, baseline PRL and GH secretion were changed in proportion to the relative intracellular hormone content. The responsiveness to TRH, KCl, and TPA during the initial 10-min period (first phase) was reduced; however, the responsiveness to these substances in the subsequent 50-min period (second phase) was unchanged. The ED50 for TRH stimulation of hormone release was increased 2- to 4-fold in down-regulated cells, but the dose-response curves for other secretagogues were not shifted. These data suggest that the initial burst of hormone release caused by TRH is mediated by Ca2+, and that prolonged exposure to TRH causes homologous desensitization.

Characterization of thyroid hormone stimulation of uridine uptake by rat pituitary tumor cells.

T3 caused a dose-related increase in the rate of [3H]uridine uptake into GH4C1 rat pituitary tumor cells. T3 increased uridine uptake to 130-180% of the control value, with a half-maximal effect at approximately 1 nM. T3 exerted a half-maximal effect at 1 h and a maximal effect at 2 h. In contrast, epidermal growth factor also increased uridine uptake by 75%, with an ED50 of 0.6 ng/ml (0.1 nM), but a half-maximal response required 4 min and a maximal effect required 20 min. T3 increased the rate of uptake at all uridine concentrations from 30 nM to 130 microM. Equilibrium binding of [125I]T3 to nuclear receptors required from 15 min at 50 nM [125I]T3 to 1 h at 0.5 nM, indicating that occupancy of nuclear receptors precedes maximal stimulation of uridine uptake. T3 did not stimulate the rate of uridine uptake at 20 C, when binding to nuclear receptors does not occur. Various thyroid hormones caused an increase in uridine uptake, with the rank order of potency 3,3',5-triiodothyroacetic acid greater than T3 greater than L-T4 greater than D-T4 approximately equal to 3,3',5,5'-tetraiodothyroacetic acid; rT3 was inactive. This order parallels the affinities of these compounds for nuclear thyroid hormone receptors.

Thyrotropin-releasing hormone rapidly stimulates a biphasic secretion of prolactin and growth hormone in GH4C1 rat pituitary tumor cells.

The characteristics of TRH-induced acute PRL and GH secretion were studied in GH4C1 cells, a clonal rat anterior pituitary tumor cell line which secretes PRL and GH. The experiments were carried out both in a flow system in which microcarrier (Cytodex)-attached cells were perifused at a constant rate and in a conventional static culture system. In both systems, cells responded to TRH in a qualitatively similar manner. TRH significantly stimulated PRL and GH secretion within 5 sec without a detectable lag period. The secretion rate was highest during the initial 1 min, declined sharply thereafter despite the continuous presence of TRH, and plateaued at a lower level. The maximum dose of TRH caused 250-700% of basal secretion during the early period (approximately 8 min; first phase) and about 150% of basal secretion thereafter (second phase). The sustained lower secretion (second phase) was maintained as long as cells were exposed to TRH (up to 2.5 h), and the secretion rate returned to the basal level within 30 min of removal of TRH from the medium. The half-maximal doses for the first and second phase secretion were 2-3 and 0.5-1 nM, respectively, in both the perifusion and static culture systems. Over a 2-day period, TRH stimulated PRL synthesis and inhibited GH synthesis. The dose-response curves for these long term effects on hormone synthesis were similar to the dose-response curves for the first phase of release. [N3-methyl-His2]TRH gave similar results, but was more potent than TRH. [N3-methyl-His2]TRH stimulated first phase release with an ED50 of 0.4-0.8 nM, second phase release with an ED50 of 0.1-0.2 nM, and hormone synthesis with an ED50 of 0.7-0.8 nM. Preincubation of the cells with Ca+2-free medium significantly depressed both first and second phase secretion. Preexposure of the cells to cycloheximide (10 micrograms/ml) had little effect on the first phase of secretion, but reduced second phase secretion. The acute effects of TRH on GH and PRL were identical, except that the secretory response tended to be greater for PRL. We conclude that 1) TRH causes hormone secretion very rapidly in a biphasic manner; 2) the first phase of secretion consists primarily of the release of stored hormone, whereas the second phase includes the release of newly synthesized hormone; 3) the dose-response curve of second phase secretion is shifted to the left compared with that of first phase secretion; and 4) both phases of secretion are at least partially dependent on extracellular Ca+2.

Characterization of multidrug-resistant pituitary tumor cells.

These studies were designed to investigate the role of P-glycoprotein in an endocrine cell line. Drug-resistant pituitary cells were obtained by growing GH4C1 cells in the presence of increasing concentrations of colchicine. Cells resistant to colchicine at 0.4 micrograms/ml, termed GH4C1/RC.4, exhibited the multidrug-resistance phenotype, as the LD50 values for colchicine, puromycin, actinomycin D, and doxorubicin were between 8 and 30 times greater than the corresponding values for the parental GH4C1 cells. Verapamil at 10 microM increased the sensitivity of GH4C1/RC.4 cells to colchicine, puromycin, and actinomycin D, almost completely reversing the drug resistance. Flow cytometry and fluorescence microscopy were used to demonstrate that GH4C1/RC.4 cells retained less rhodamine 123 than GH4C1 cells, and that the rate of efflux of rhodamine 123 was much faster for GH4C1/RC.4 cells. Immunocytochemical staining with a monoclonal antibody, C219, to the 170-kilodalton P-glycoprotein showed directly that GH4C1/RC.4 cells overexpress P-glycoprotein. We used drug-resistant pituitary cells to assess the possible role of P-glycoprotein in uptake and efflux of several hormones. At equilibrium, GH4C1 and GH4C1/RC.4 cells bound similar amounts of [125I]L-triiodothyronine and [125I]L-thyroxine, and verapamil did not alter either equilibrium binding or thyroid hormone efflux kinetics. Multidrug-resistant GH4C1/RC.4 cells retained less [3H]hydrocortisone than parental GH4C1 cells at equilibrium, and verapamil increased the equilibrium concentration of [3H]hydrocortisone 3.6-fold. The effect of verapamil was due to its ability to reverse multidrug resistance, since two other chemosensitizers, quinidine and vinblastine, increased [3H]hydrocortisone retention as effectively as verapamil but another calcium channel blocker, nifedipine, had no effect. The drug-resistant GH4C1/RC.4 line synthesized more GH (290%) and much less PRL (5%) than the parent. Hydrocortisone stimulated GH synthesis and inhibited PRL similarly in GH4C1 and GH4C1/RC.4 cells. The results show that the GH4C1/RC.4 line is multidrug-resistant and overexpresses the 170-kilodalton P-glycoprotein and suggest that the P-glycoprotein pump contributes to hydrocortisone kinetics.

Degradation of thyrotropin-releasing hormone by the GH3 strain of pituitary cells in culture.

Thyrotropin-releasing hormone (TRH), pGlu-His-ProNH2, binds within 1 h to specific receptors on the GH3 strain of pituitary cells. When GH3 cells were incubated for 2 days with 3 nM [2,3-3H-Pro]TRH, an increasing fraction of the total cellular radioactivity (7% after 1 h, 81% after 43 h) was associated covalently with proteins as determined by dialysis, acid precipitation, and gel filtration; this fraction corresponded to label which could not be displaced from intact GH3 cells by the addition of excess unlabeled TRH. R5 and GH12C1 cells, strains which lack TRH receptors, accumulated 16 or 23%, respectively, as much label from [2,3-3H-Pro]TRH as did GH3 cells in 24 h. After 24 h of incubation with [2,3-3H-Pro]TRH and [14C-His]TRH, the ratio of 14C/3H in GH3 cells was the same as in the culture medium, indicating that the intact tripeptide was taken up by the cells. After 24-48 h of incubation with [2,3-3H-Pro]TRH, GH3 proteins appeared to be labeled randomly as surmised by fractionation on Sephadex G-100, DEAE cellulose, Sepharose 4B and sucrose density gradients. In cultures treated with cycloheximide (10 mug/ml) or proline (6.3 mM) the initial binding of [2,3-3H-Pro]TRH to receptors, measured after 1 h, was 97% or 102% of control. However, the incorporation of label from [2,3-3H-Pro]TRH into an acid-precipitable product after 22 h was inhibited by 81 and 74% by cycloheximide (1 mug/ml) and proline (2.5 mM). Formation of [2,3-3H] proline from [2,3-?3H-Pro] TRH was demonstrated by thin layer chromatography; the percentage of non-protein radioactivity with an Rf of proline increased from 20 to 80% in GH3 cells incubated 1 or 24 h with [2,3-3H-Pro]TRH. We conclude that after binding to receptors on GH3 cells, TRH is slowly metabolized to its constituent amino acids, and the products [2,3-3H]proline or [14C]histidine are incorporated into newly synthesized proteins.

Thyrotropin-releasing hormone-induced intracellular calcium responses in individual rat lactotrophs and thyrotrophs.

Calcium responses to TRH were recorded for individual cells cultured from rat anterior pituitary tissue loaded with fura-2, and cell type was subsequently identified by immunocytochemistry. At 100 nM and 1 microM, TRH stimulated a single transient spike of intracellular free calcium ([Ca2+]i) in 95-100% of lactotrophs. At a concentration of 10 nM or less, the proportion of TRH-responsive cells decreased, and the [Ca2+]i responses became more heterogeneous, consisting of a biphasic response in which an initial [Ca2+]i spike was followed by a sustained elevation of [Ca2+]i or [Ca2+]i oscillations. Initiation of TRH-induced oscillations required the release of intracellular Ca2+ from thapsigargin-sensitive stores, whereas maintenance of the oscillations required influx of extracellular Ca2+ through nimodipine-sensitive Ca2+ channels. The amplitude of the initial [Ca2+]i rise increased from 0.1-10 nM TRH and was not significantly reduced by removal of extracellular Ca2+. The duration of the initial [Ca2+]i transient was significantly shorter at 1 microM than at 1 nM TRH. When TRH was added to cells that had been treated with thapsigargin to block the agonist-induced [Ca2+]i increase, TRH often decreased [Ca2+]i, particularly in cells with high [Ca2+]i. These results suggest that TRH and elevated [Ca2+]i act as coactivators of Ca2+ efflux, which helps terminate the agonist-evoked [Ca2+]i transient. In addition, TRH caused increases in [Ca2+]i in individual rat thyrotrophs, and these responses were heterogeneous. TRH stimulated a [Ca2+]i response in a lesser proportion of thyrotrophs from euthyroid compared to hypothyroid male rats. Essentially all TRH-responsive cells stained for either PRL or TSH.

Characteristics of the Ca2+ spike and oscillations induced by different doses of thyrotropin-releasing hormone (TRH) in individual pituitary cells and nonexcitable cells transfected with TRH receptor complementary deoxyribonucleic acid.

The Ca2+ response of individual cells to TRH was investigated in excitable pituitary GH3 and in nonexcitable Hela cells transfected with the TRH receptor complementary DNA (HelaR cells). GH3 cells typically responded to 1 microM TRH with an immediate transient [Ca2+]i spike (mean peak [Ca2+]i = 1.5 microM) followed by a period of inactivity of approximately 100 sec long and then a secondary increase in [Ca2+]i with oscillations. At 10-100 nM TRH, the initial [Ca2+]i spike was more prolonged and immediately followed by a sustained elevation of [Ca2+]i. At 0.5-1 nM TRH, there was a variable lag before any response; the initial [Ca2+]i spike was absent or small, but the sustained phase was still present. The second phase of elevated [Ca2+]i, which could be eliminated with nimodipine or chelation of extracellular Ca2+, gave a bell-shaped TRH dose response curve. The effect of TRH on Ca2+ oscillations depended both on TRH concentration and the basal oscillation frequency. HelaR cells responded to 1 microM TRH with a rapid [Ca2+]i spike, and at less than or equal to 10 nM TRH, up to 50% of HelaR cells displayed agonist-induced sinusoidal [Ca2+]i oscillations independent of extracellular Ca2+. TRH never caused a sustained elevation of [Ca2+]i in HelaR cells. For GH3 and HelaR cells, the peak [Ca2+]i response increased with TRH concentration up to 1 microM. In contrast, the duration of the initial [Ca2+]i spike was shorter at higher TRH concentrations, decreasing from 16 to 6.3 s (GH3) or 92 to 35 s (HelaR) between 0.5 nM and 1 microM TRH. This shortening of the spike duration was caused by rapid clearing of cytoplasmic Ca2+ that depended primarily on agonist concentration. In summary, TRH stimulates a complex [Ca2+]i response pattern dependent upon both the agonist concentration and cell context. The initial burst of Ca2+ is cleared in part by agonist dependent Ca2+ clearing.