Thyrotropin-releasing Hormone Increases Cytosolic Free Ca" in Clonal Pituitary Cells (GH

Changes in the cytosolic free Ca" concentration following cell surface receptor activation have been proposed to mediate a wide variety of cellular responses. Using the specific Ca2+ chelator quint as a fluorescent intracellular probe, we measured the Ca" levels in the cytosol of clonal rat pituitary cells, GH 3 cells . We demonstrate that thyrotropin-releasing hormone (TRH) at nanomolarconcentrations leads to a rapid and transient increase in cytosolic Ca2+ . This increase was found to occur in Ca 2'-free media in the presence of EGTA, thus at extracellular Ca2+ levels that are below the cytosolic concentrations, and was not prevented by verapamil, a Ca2+ channel blocker. Depolarization of GH 3 cells with K+, which can mimic the action of TRH on prolactin release, increased cytosolic Ca2+ levels only in the presence of free extracellular Ca2+, and this increase could be blocked by verapamil . These data show that the mobilization of intracellular Ca2+ due to TRH action that has been proposed by previous studies actually leads to an increase in cytosolic free Ca 2' . The kinetic features of this response emphasize the key role of cytosolic free Ca2+ in stimulus-secretion coupling . Thyrotropin-releasing hormone (TRH)' stimulates the secretion of prolactin and thyrotropin from the pituitary. Changes in intracellular Ca 2' distribution have been implicated in the mediation of TRH action (for review see reference 1) . TRH increases the frequency of Ca2' action potentials in pituitary cells (2) or in GH3 cells (3, 4), a clonal rat pituitary cell line responsive to TRH (5) . Depolarization of GH3 cells with K+, which promotes the entry of Ca2+' can mimic some of the actions of TRH (6) . It has been proposed that TRH increases influx of extracellular Ca2+ (2-4, 6, 7), but it has also been demonstrated that TRH leads to an increased Ca2' efflux from the cells (8-10) and can cause a net loss of Caz+ from GH3 cells (10) . The use ofdigitonin made it possible to reveal distinct pools of intracellular Ca2' and it was reported that TRH can lead to a rapid depletion of Ca2+ from two pools that have been tentatively identified as of the mitochondria 'Abbreviations used in this paper: bis-oxonol, bis-(1,3-diethylthiobarbiturate)-trimethineoxonol ; [Ca2+1,' cytosolic free Caz+ concentration ; KRBH, HEPES-buffered Krebs-Ringer bicarbonate bufferquint/AM, quint acetoxy methyl ester ; TRH, thyrotropin-releasing hormone . THE JOURNAL OF CELL BIOLOGY " VOLUME 99 JULY 1984 83-87 C The Rockefeller University Press 0021-9525/84/07/0083/06 $1 .00 and of the endoplasmatic reticulum (11) . However, none of the studies so far could give direct information on a presumed key variable in stimulus-secretion coupling, namely, the concentration of free cytosolic Ca 2+' [Ca2+],. To measure the effects of TRH on [Caz+] ; we used the fluorescent probe quint, which was introduced by Tsien (12) and has been applied to lymphocytes (13, 14), platelets (15), neutrophils (16), and an insulin-producing cell line (17) . quint can be introduced into cells as the lipophilic acetoxy methyl ester (quint/AM) which is hydrolyzed by cellular esterases to yield hydrophilic free quint . Its fluorescence increases in the presence ofCa2+ which is complexed selectively and with little interference from either Mgz+ or protons . The apparent dissociation constant for the fluorescent Ca" quint complex in the presence of 1 MM Mgz+ (the presumed cytoplasmic Mgz+ concentration) is 115 nM. Since this is close to cytoplasmic levels of Ca2+ in many cells (18, 19), the fluorescent probe monitors even relatively small variations in [Caz +] ; . The intracellular quint fluorescence can be calibrated, which allows a quantitative measure of [Ca 2+], . This present paper presents evidence, obtained with this technique, that TRH transiently increases [Ca2+], in both the presence and absence of extracellular Caz+ . 83 on S etem er 9, 2017 jcb.rress.org D ow nladed fom MATERIALS AND METHODS Cell Culture : GH3 cells obtained from the American Type Culture Collection (Rockville, MD) were used throughout . Stock monolayer cultures wre maintained as described (5). Before the experiment the cells were detached from the culture flasks with 0.02% EDTA and incubated in HEPES-buffered (30 mM, pH 7 .4) Ham-1710 medium supplemented with 15% horse and 2.5% fetal calfserum in spinner flasks at 37°C (1) for 3 h or overnight. Determination of [Ca 2+], in GH3 Cells with the Fluorescent Intracellular Probe quint : The validity of the quint method for assessing [Ca2 'J; has been verified in detail with lymphocytes (13, 14) . The experimental conditions for the application of this method to GH3 cells were verified to assure a sufficient level of intracellular probe, the unchanged spectral characteristics of intracellular quint fluorescence (see Fig. 1) indicating complete hydrolysis of quint/AM and negligible quenching of the signal, the cytosolic location of quint, and the maintenance of cellular functions. Before being loaded, the cells were centrifuged twice (100 g for 10 min) and resuspended in culture medium RPMI 1640 containing 25 mM HEPES (pH 7 .4) and 0.5% BSA (fraction V) . The final cell concentration was 20-30 x 106 cells/ml. quint/AM (Lancaster Synthesis, Morecambe, Lancashire, U.K.) was added to the cell suspension from a 10-mM stock solution in DMSO to a final concentration of 100 pM . After 10 min at 37*C, 4 vol of RPMI 1640 were added and the incubation was continued for 50 min . Control cells were incubated in parallel with 1 % DMSO and then diluted as the test cells . After the incubation all cells were washed and resuspended in RPMI 1640 without serum or albumin and kept at room temperature until use. Fluorescence of quint external standard and loaded cells was measured in 1-cm glass cuvettes thermostatted at 37°C in a Perkin-Elmer fluorescence spectrophotometer (LS-3 ; Perkin-Elmer Corp ., Eden Prairie, MN) at wavelengths 339 nm for excitation and 492 nm for emission. Membrane potential was recorded with the fluorescent probe bis-(1,3-diethylthiobarbiturate)-trimethineoxonol (bis-oxonol), a lipophilic anion (20) . Bis-oxonol fluorescence was measured at excitation and emission wavelengths of 540 and 580 nm, respectively . The slit widths were set to cover a spectral range of 10 min (excitation) and 20 nm (emission), for both quint and bis-oxonol measurements . The cells (5 x 105-2 .5 x 106 cells/ml) were suspended in 2 ml of a modified KrebsRinger bicarbonate buffer containing 25 mM HEPES, 5 mM NaHC03, 1 .1 mM CaC12, 0 .1 mM EGTA, and 2.8 mM glucose (KRBH), and were stirred continuously with a magnetic stirrer. Test agents were addedfrom concentrated stock solutions. The fluorescence signal was calibrated at the end ofeach individual trace as follows : To obtain the minimal fluorescence signal of quint, 4 mM EGTA (pH 7 .4) and 30 mM Tris were added to yield an approximate free [Cat '] of 1 nM. These additions show the level of extracellular quint (13, 14), as the decrease of the extracellular [Cat'] leads to a rapid (<10 s) initial decrease of the fluorescence of the extracellular dye . At this stage the hydrolysis ofquint/AM was usually verified by recording the emission spectrum (Fig . 1) . After the cells were lysed with Triton X-100 (final concentration 0 .1 %), all the quint was exposed to [Ca t'] of -1 nM and the minimal fluorescence signal, F~i , was obtained . The maximal quint fluorescence, F_was then determined by readdingCa t' (4 mM) to the cuvette restoring a free [Ca t'] of -1 mM . For quint measurements performed in Cat '-free KRBH, the level of extracellular quint was determined by restoring the free [Cat '] to 1 mM before following the procedure outlined above which was then used to obtain F.I and F . The difference between F.; andF.., corrected for the dilution occurring by the addition of the various reagents, also comprises the extracellular quint fluorescence which has to be subtracted to obtain the maximal fluorescence due to intracellularly trapped quint, 17 ,4 * ; F -* averages 71 ± 5% of Fmu (± SEM, n = 10). To establish the scale of [Cat'] relative to the fluorescence trace, Fm; was corrected for the changes in fluorescence due to the detergent that were measured in nonloaded cells in parallel, and, in the case of Ca"-containing KRBH, also for the contribution of extracellular quint, yielding Fmi *. [Cat']i can then be calculated from the fluorescence F at each point of the trace as [Ca t'] ; = Kd(FFm; *)/(Fmu* F), and Ka = 115 nM (for details see reference 13). The intracellular concentration of quint was calculated by comparing F_* to the fluorescence of a known quantity of quint added to control cells incubated under the conditions described to establish Fu . quint loading of GH 3 cells averaged 0.46 ± 0.07 mM (± SEM, n = 11) calculated with an intracellular volume of 1 .4 1A1/106 cells (21); there was no apparent correlation between quint loading ranging from 0.24 to 0.9 mM and prestimulatory [Ca"II . The cytosolic localization of quint in GH 3 cells was confirmed by the correlation between the release of quint and the release of lactate dehydrogenase, a cytosolic enzyme, upon digitonin treatment. quint-loaded GH3 cells were exposed to increased concentrations of digitonin (2 .5 MM to 1 mM) in Ca"free KRBH, 1 mM EGTA, for 5 min at 0°C; in the supernatant after centrifuH4 THE JOURNAL OF CELL BIOLOGY VOLUME 99, 1984 gation (1500 g, 5 min), quint was determined by fluorescence and lactate dehydrogenase was measured following the change in absorbance at 340 nM with NADH and pyruvate. 85-90% of quint and >95% of lactate dehydrogenase were released at maximal digitonin levels, and for intermediate digitonin concentrations there was a strict correlation (R = 0 .98, n = 16) between the release of quint and lactate dehydrogenase. ATP levels were determined after extraction of GH3 cells with 6% HCI04 and neutralization, by high-performance liquid chromatography on a Partisil 10 SAX ion exchange column (4 .6 x 25 cm) eluted isocratically with 0.5 M ammonium phosphate, pH 4 .4, at a flow rate of 2 ml/min. It was found that quint loading led to a slight but not significant increase of the ATP levels in GH 3 cells (1 .48 ± 0.13 [± SEM, n = 41 versus 1 .24 ± 0 .19 nmol/106 cells) . From these experiments we conclude that the quint loading does not cause a major disturbance of cellular homeostasis .