Regulation by Calcium of Prolactin and Growth Hormone mRNA Sequences in Primary Cultures of Rat Pituitary Cells*

Addition of Ca2+ to primary cultures of female pitui- tary cells incubated in serum-free medium lacking added Ca2+ yielded no effects on levels of prolactin or growth hormone mRNA, assayed by cytoplasmic dot hybridization. However, incubation of the cells in serum-free medium containing sufficient ethylene glycol bis(@-aminoethyl ether)-N,N,N',N'-tetraacetic acid to reduce medium Ca2+ levels below the 10-40 WM present as a trace contaminant yielded a decrease in the levels of both mRNAs. The decrease was dose- dependent at extracellular Ca2+ concentrations below 1.0 PM, had an apparent half-maximum at about 0.3 p ~ , and did not appear to plateau with increasing incubation times. Following 2-3-day incubations of cells in low Ca2+, a reduction of prolactin mRNA (23-70-fold) consistently greater than the reduction of growth hormone mRNA (9-15-fold) was observed. Similar effects of reduced extracellular Ca2+ were obtained with primary cultures of male pituitary cells. The specificity of these effects of lowered extracellular Ca2+ was by observations.

Addition of Ca2+ to primary cultures of female pituitary cells incubated in serum-free medium lacking added Ca2+ yielded no effects on levels of prolactin or growth hormone mRNA, assayed by cytoplasmic dot hybridization. However, incubation of the cells in serum-free medium containing sufficient ethylene glycol bis(@-aminoethyl ether)-N,N,N',N'-tetraacetic acid to reduce medium Ca2+ levels below the 10-40 WM present as a trace contaminant yielded a decrease in the levels of both mRNAs. The decrease was dosedependent at extracellular Ca2+ concentrations below 1.0 PM, had an apparent half-maximum at about 0.3 p~, and did not appear to plateau with increasing incubation times. Following 2-3-day incubations of cells in low Ca2+, a reduction of prolactin mRNA (23-70-fold) consistently greater than the reduction of growth hormone mRNA (9-15-fold) was observed. Similar effects of reduced extracellular Ca2+ were obtained with primary cultures of male pituitary cells. The specificity of these effects of lowered extracellular Ca2+ was demonstrated by the following observations. The decreases in these mRNAs were substantially reversible by readdition of Ca2+ to the incubation medium. Reduction of extracellular Ca2+ led to no detectable changes in cellular ribosomal RNA levels or overall RNA synthesis. In male pituitary cells, the level of another metal-regulated mRNA, that for metallothionein, was not decreased by a reduction of extracellular Ca2+ that caused a 40-fold decrease in levels of prolactin and growth hormone mRNA. Hence, Ca2+ exhibits specificity in its regulation of pituitary prolactin and growth hormone gene expression.
The role of free intracellular Ca2+ in the regulation of a host of cellular processes, including adenohypophyseal hormone secretion, is currently an area of intense investigation (1). By contrast, there have been few studies of the possible involvement of Ca2+ in the regulation of specific gene expression. In previous studies with a line of rat pituitary (GH,) cells, we showed that addition of Ca2+ to cells which had been incubated in a Ca'+-depleted medium strongly stimulates (7-200-fold) prolactin mRNA levels, while only slightly increas- ing (by 60%) growth hormone mRNA levels (2). Combinatorial analysis under these conditions of the regulation by epidermal growth factor, thyrotropin-releasing hormone, and/ or Ca2+ suggested a possible role for Ca2+ as a mediator of peptide hormone action on prolactin gene expression (3).
It seemed possible that the results of the Ca2+ studies described above arose from the transformed nature of the GH3 cells and thus might not represent an accurate reflection of regulatory mechanisms occurring in normal rat pituitary cells. To investigate this possibility, we have carried out studies of the regulation by Ca2+ of prolactin and growth hormone mRNA levels in primary cultures of normal rat pituitary cells. We report here that prolactin and growth hormone mRNAs are regulated in a qualitatively similar fashion in normal pituitary cells and the GH3 cells. However, the effects of Ca2+ occur over a much lower extracellular concentration range in pituitary cells than in the GH3 cells.

EXPERIMENTAL PROCEDURES
Cell Culture-A monolayer serum-free medium (SFM') containing no added Ca2+ or hormones, prepared as described previously (2), was employed for all incubations. Primary cultures of rat pituitary cells were prepared by trypsin digestion essentially as described by Hymer et al. (4). Pituitaries removed from either female retired breeder or male Sprague-Dawley rats were minced in SFM to pieces approximately 1 mm in size. The pieces were washed twice and incubated a t 37 "C in SFM plus trypsin and bovine serum albumin (1 mg/ml each; Sigma) for 2 h, during which the pieces were repipetted at 30-min intervals. The resultant dissociated cells were then diluted 10-20-fold with SFM and plated a t 0.5 X 106/ml in 35-mm plastic Petri dishes (not tissue culture-treated). The average cell yield was 1 X lo6 and 2-3 X lo6 from individual male and female pituitaries, respectively. Cell viability, determined by trypan blue exclusion, was greater than 99%. Cells were incubated a t 37 "C in 5% COn and were observed to attach to the plates within 15 min. The medium was changed after 15 min and about 16 h, a t which time experimental additions were made.
Determination of eaZ+ Concentration of Serum-free Medium-A spectrophotometric assay employing the metallochromic Ca" indicator arsenazo 111 (0-[1,8-dihydroxy-3,6-disulfonaphthalene-2,7bis(azo)]bisbenzenearsonic acid) (5) was employed. Arsenazo 111 binds free Ca2+ avidly and is most sensitive to Ca2+ in the micromolar range. A standard curve was constructed by adding known amounts of Ca2+ or EGTA to SFM. One-half of each sample was made 1.0 mM in EGTA; arsenazo I11 (final concentration 100 WM) was added to all samples, and the absorbance difference a t 650 nM between samples incubated plus or minus EGTA was measured in a dual-beam spectrophotometer. The Ca2+ concentration of individual preparations of SFM ranged from 10 to 40 PM. A value for the KO of CaZ+-EGTA of 0.15 WM (6) was employed to calculate the amount of EGTA required to yield the desired SFM Ca2+ concentration.
Quantitation of Specific RNAs-Cells were scraped with a rubber policeman into phosphate-buffered salts. Relative levels of prolactin or growth hormone mRNA, or 18 S ribosomal RNA were then assayed by cytoplasmic dot hybridization as described (7). Briefly, cytoplasmic samples were heated in the presence of formaldehyde, 2-fold dilutions dotted onto each of three nitrocellulose sheets, and the sheets baked. Each sheet was then hybridized against a 3ZP-labeled hybrid plasmid (2-4 X 108cpm/pg, IO7 cpm/lO ml) containing cDNA for rat prolactin (pPRL-1 (8)), rat growth hormone (pBR322-GH1 (9)), or 18 S ribosomal DNA (pXC-1 (lo)), washed, autoradiographed, and scanned. When a wide range of spot intensities was analyzed, a series of autoradiographic exposures was employed to yield conditions where the intensity of each spot was in the linear range of the film. All values of mRNA levels have been divided by 18 S ribosomal RNA content, which was little affected by variations in extracellular Ca2+ (see "Results").
Total RNA Synthesis-Following incubation in 2.0 ml of SFM containing 20 pCi of [3H]uridine for 30 min, cells were scraped into phosphate-buffered salts, washed once, and lysed by vortexing 30 s in 0.1% sodium dodecyl sulfate, 150 mM NaCl, 1 mM EDTA, 10 mM Tris, pH 7.5. Trichloroacetic acid-precipitable radioactivity was determined by scintillation counting.

Ca2+ Regulation of Prolactin and Growth Hormone mRNA Levels in Female Pituitary
Cells-In previous experiments, it was found that addition of Ca2+ (0.4 mM) to a line of rat pituitary tumor (GH,) cells incubated in a serum-free medium lacking added Ca2+ yielded a maximal and sizable stimulation of prolactin mRNA levels (2). However, in preliminary experiments, when Ca2+ was added at concentrations up to 1.0 mM to primary cultures of female rat pituitary cells incubated in the same serum-free medium, no significant stimulation of cytoplasmic levels of either of these mRNAs occurred (data not shown).
Assays with arsenazo I11 of the Ca'+ concentration of the serum-free medium employed in these experiments yielded values in the range of 10-40 p~. It seemed possible that normal pituitary cells incubated at these low levels of extracellular Ca'+ could maintain high intracellular Ca2+ levels, thus preventing a response to exogenously added Ca2+. The Ca2+ concentration of this medium was therefore further reduced by the addition of EGTA. Pituitary cells were incubated for 72 h in the presence of a range of resultant Ca'+ concentrations, and the effect on levels of prolactin mRNA, growth hormone mRNA, and ribosomal RNA was analyzed by cytoplasmic dot hybridization (Fig. 1). Decreasing extracellular Ca2+ from 25 to 5 p~ yielded no detectable effect on cellular levels of either prolactin or growth hormone mRNA. However, incubation at 0.1 p~ Ca'+ significantly reduced the levels of both mRNAs. This reduction was not due simply to an overall change in the RNA metabolism of these cells, since incubation for 72 h at this Ca'+ concentration led to no detectable change in ribosomal RNA levels (Fig. l), nor in overall RNA synthesis (data not shown). The observation that addition of Ca2+ but not M$+ to the incubation medium reversed the effects of EGTA shows that these effects are due to chelation by EGTA of Ca'+. Finally, it can be seen in Fig.  1 that female pituitaries incubated with high (5-25 p~) Ca2+ contain considerably more prolactin mRNA than growth hormone mRNA. The dose-response relationship for prolactin and growth hormone mRNA were similar, with apparent half-maximal effects at about 0.3 p~ Ca'+ (Fig. 2). However, at the lowest prolactin and growth hormone mRNA levels were reduced to 1.4 and 6.7%, respectively, of the levels observed in cells incubated at 1.0 pM Caz+. Thus, prolactin mRNA levels appear to be more sensitive to very low extracellular Ca2+ than are growth hormone mRNA levels (see also Fig. 3 and Table I). As observed previously, total RNA synthesis was constant over the range of Ca2+ concentrations tested (Fig. 2, bottom panel).

Dose Dependence of Prolactin and Growth Hormone mRNA on Extracellular
Kinetics The effects on prolactin and growth hormone mRNA levels and on overall RNA synthesis are shown in Fig. 3. Incubation for 6 days in high Ca2+ yielded no detectable effect on prolactin mRNA levels and a gradual decrease in growth hormone mRNA levels to about 50% of their initial value. Incubation for 1 day in low Ca2+ yielded no effect on prolactin mRNA levels and a small (30%) decrease in growth hormone mRNA levels. Between 1 and 3 days, in the presence of low Ca", levels of both mRNAs declined with approximately first-order  Growth Hormone mRNA 7617 kinetics, with half-times for prolactin mRNA and growth hormone mRNA of approximately 12 and 19 h, respectively. By 6 days of incubation with low Ca2+, prolactin mRNA and growth hormone mRNA levels had declined by approximately 80-and 35-fold, respectively, relative to the levels in cells incubated with high Ca2+. During the 6-day incubation period, overall RNA synthesis by the cells declined gradually to about 50% of its initial value, with no apparent difference at any time point between the cultures incubated in the presence of high or low Ca2+ (Fig. 3, bottom panel).
Reversibility of the Effect of Reduced Extracellular Ca2+-It seemed possible that the decrease in prolactin and growth hormone mRNA levels in pituitary cells incubated in the presence of low extracellular Ca2+ was due to irreversible inactivation of the pituitary cells which produce these mRNAs. T o investigate this possibility, pituitary cells were incubated in the presence or absence of EGTA (20 FM) for 3 days, after which some of the EGTA-treated cultures received Ca2+ (0.4 mM). It is seen in Table I that incubation of pituitary cells with EGTA for 6 days caused prolactin and growth hormone mRNA levels to decrease to 4 and 11%, respectively, of the levels in control cultures incubated without EGTA. Addition of Ca2+ after 3 days to the EGTA-treated cultures led, by 6 days, to increases of both mRNAs to about 30-35% of the levels in the control cultures. Thus, the effects of incubation in low Ca2+ on the levels of both mRNAs are substantially reversed by readdition of Ca2+ to the incubation medium. As observed previously, none of these incubation conditions led to significant changes in either total RNA synthesis or ribosomal RNA content (Table I).

I n Male Pituitary Cells ea2+ Regulates the mRNAs for
Prolactin and Growth Hormone, but Not Metallothionein mRNA-To investigate whether the effects described above are sex-dependent, the effect of Ca2+ on mRNA levels in male pituitary cells was investigated (Fig. 4). It is seen in Fig. 4A that Ca2+ exerted effects quite similar to those in female cells. No effect was observed as extracellular Ca2+ was reduced from 10 to 0.4-0.6 FM. As the Ca2+ concentration was further reduced, cellular levels of prolactin and growth hormone mRNA declined sharply.
The observation (Figs. 1-3 and Table I) that lowering extracellular Ca2+ had no effect on pituitary cell total RNA synthesis or ribosomal RNA content showed that the accompanying effects on prolactin and growth hormone mRNA were not due simply to a change in the overall RNA metabolism of the cells. However, lowering extracellular Ca" caused similar declines in both prolactin and growth hormone mRNA. Hence, to examine further the specificity of the effect on the levels of these two mRNAs, the influence of Ca2+ on expression of another metal-regulated gene was investigated. Metallothionein gene expression in many cell types is stimulated by heavy metals (11). Reprobing one of the nitrocellulose sheets employed for the analysis shown in Fig. 4A with a rat metallothionein cDNA probe yielded the results shown in Fig.  4B. When extracellular Ca2+ concentrations were lowered, metallothionein mRNA levels did not decrease. Rather, levels of this mRNA were stimulated by decreasing extracellular ca2+ in the range from 10 to 0.6 PM. When extracellular Ca2+ was reduced from 0.6 to 0.15 PM, there was a 40-fold decrease in prolactin and growth hormone mRNA levels, while metallothionein mRNA levels were not significantly affected.

DISCUSSION
We have found that incubation of rat pituitary cells in a serum-free medium containing submicromolar Ca2+ concen- trations leads to sizable decreases in cellular levels of both prolactin and growth hormone mRNA. However, incubation of the cells under these conditions did not affect either total RNA synthesis or ribosomal RNA content and did not reduce cellular levels of metallothionein mRNA. Thus, Ca2+ depletion exhibits specificity in its ability to reduce expression of the prolactin and growth hormone genes.
Incubation of female pituitary cells for 2-3 days at submicromolar Ca2+ concentrations consistently yielded a greater maximal reduction, relative to cells incubated at micromolar Ca2+ concentrations, of prolactin mRNA (23-70-fold) than of growth hormone mRNA (9-15-fold). These observations agree qualitatively with previous results obtained with GH3 cells, in which addition of Ca2+ caused a large (7-200-fold) stimulation of prolactin mRNA (2), and a small (1.6-2.0 fold) stimulation of growth hormone mRNA (2,12). However, the dose dependence of the Ca2+ effects in GH3 cells and normal rat pituitary cells are quite different. Thus, GH3 cells exhibited a halfmaximal stimulation of prolactin synthesis at 70 FM extracellular Ca2+ (2), while in primary cultures of normal rat cells half-maximal reduction in either prolactin or growth hormone mRNA was not achieved until extracellular Ca2+ concentrations of about 0.3 WM were reached (Fig. 2). It seems likely that the basis for this quantitative difference arises from an ability of normal prolactin-and growth hormone-producing pituitary cells to maintain a basal cytosolic free intracellular Ca2+ concentration ([Ca2+Ii) in the presence of extremely low extracellular Ca2+. In GH, cells, [Ca2+Ii has been reported to be 0.12 (13) or 0.35 PM (14), i.e. in the range of the extracellular Ca2+ concentrations yielding half-maximal effects in the Incubation of pituitary cell cultures for 6 days in serumfree medium containing a high (25 PM) Ca2+ concentration yielded no detectable change in cellular levels of prolactin mRNA (Fig. 3). This agrees with previous observations which showed that prolactin synthesis is unaffected by prolonged incubation of pituitary cells in a serum-free medium (16). Since no hormones or undefined components were present in the medium employed in the present studies, this observation suggests that Ca2+ alone may be sufficient to maintain basal prolactin gene expression. By contrast, under the same incubation conditions, growth hormone mRNA was observed to decrease gradually with time (Fig. 3). The significance of this observation is not entirely clear, since under these conditions overall RNA synthesis decreased at about the same rate (Fig.   3). However, this result suggests that basal growth hormone gene expression is dependent upon other factors beside Ca'+.
Precise determination of the half-life of an mRNA requires the use of pulse-chase techniques. However, the decay kinetics of an mRNA following removal of an inducer can be employed to estimate the half-life of the mRNA in the absence of the inducer. Thus, the mRNA decay kinetics observed in pituitary cells incubated in medium containing low (0.1 PM) Ca2+ (Fig.  3) suggest that the half-life of the mRNAs for prolactin and growth hormone in pituitary cells incubated in the presence of low Ca2+ are approximately 12 and 19 h, respectively.
The present studies show that Ca2+ regulates expression of the prolactin and growth hormone genes a t a pretranslational level, but do not yield direct information concerning the regulation by Ca'+ of the transcription of these genes. However, we have shown previously that incubation of GH, cells with Ca'+ yields increases in all detectable nuclear prolactin mRNA precursors (3). Furthermore, nuclear run-on transcription experiments have shown that incubation of normal pituitary cells in the presence of low extracellular ca'+ causes decreases in the transcription of the prolactin and growth hormone genes which parallel the decreases in mRNA levels reported here (12). Hence, it appears that the Ca'+ effects observed in the present experiments are exerted, at least partially, at the level of transcription.
The evidence presented here for Ca'+ regulation of prolactin and growth hormone mRNA levels in the pituitary is consistent with previous reports describing a role for Ca2+ in the biosynthesis of prolactin (17) and growth hormone (17, 18).
In addition, previous investigations have demonstrated a role for Ca2+ in the secretion of prolactin (17)(18)(19) and growth hormone (17, 20, 21). The cellular mechanisms involved in the role of Ca2+ in both the expression of these genes and the secretion of their protein products remain to be elucidated. It is, however, intriguing to speculate that these apparently disparate cellular processes may be functionally related.