Dependence of GnRH action on Na+, K+, and Ca2+ in the frog, Rana pipiens, pituitary

THE JOURNAL OF EXPERIMENTAL ZOOLOGY 239:379-391(1986)

Dependence of GnRH Action on Na+ , K+ , and Ca2+ in the Frog, Rana pipiens, Pituitary

DAVID A. PORTER AND PAUL LICHT Department of Zoology, University of California, Berkeley, California 94720

ABSTRACT The roles of K+, Ca2+, and Naf ions in the mechanism of gonadotropin releasing hormone (GnRH) action on frog (Rana pipiens) hemipituitaries were studied using an in vitro superfusion system. The effects of elevated K+ alone or in combination with Ca2+-depleted medium, tetrodotoxin W X ) , or with 100 ng/ml GnRH were examined. The involvement of Kf was also studied indirectly through the use of tetraethyl ammonium chloride (TEA). The importance of Ca2+ was established by the loss of res onsiveness to GnRH in Ca2+-depleted medium, or in the presence of the Ca'+ competitor CoC12. The absence of a major dependence of GnRH on Na+ was revealed by the continued gonadotropin secretion after addition of 1 pM 'ITX to medium containing GnRH or 36.3 mM KC1, or by replacement of NaCL with choline chloride. High (10 x normal) KC1 (36.3 mM) stimulated gonadotropin-both LH and FSH-secretion, but the response was more gradual than for GnRH. The inclusion of TEA (to block Kf efflux) in medium with GnRH accentuated the effect of GnRH, and the effects of elevated (36.3 mM) KC1 and 100 ng/ml GnRH (a relatively high dose) were additive. Responses to high K+, like GnRH, were abolished by removal of Ca2+ from the medium. Overall, the roles of K+, Ca2+, and Naf ions in the mechanism of GnRH action are very similar between mammals and frogs; Ca2+ apparently serves a critical function in the mechanism of GnRH action, while Na+ appears not to be involved. K+ can induce gonadotropin secretion, but it is not clear that it plays a direct role in the mediation of the action of GnRH.

Pituitary glands from the frogs, Rana catesbeiana and Rana pipiens, are more resis- tant to desensitization to relatively high doses of gonadotropin-releasing hormone (GnRH) than are many mammalian pituitaries. For example, in vivo studies have indicated that R. catesbeiana retain responsiveness to continually infused GnRH for up to 4 days (McCreery and Licht, 1983a,b). Rana pipiens hemipituitaries retain responsiveness to even high doses of GnRH superfused continuously for up to 48 hr (Porter and Licht, '85). As in mammals (e.g., Pickering and Fink, '76; Liu and Jackson, '78; Bourne and Baldwin, '801, the release of the gonadotropins (GtH), luteinizing hormone (LH), and follicle-stimulating hormone (FSH) in response to GnRH is biphasic (both in vivo and in vitro). Only the release of the two GtHs during the second phase of GnRH stimulation requires protein synthesis (Porter, '85).

The phylogenetic position of the Amphibia makes the study of the mechanism of GnRH action in frogs valuable for understanding the evolution of vertebrate hormone action. However, it is not yet clear what differences in the mechanisms of GnRH action account for the differences between mammals and frogs in resistance to desensitization to chronic treatment with GnRH. To examine this problem further, this study undertook to ascertain in broad terms the ionic requirements, especially the roles of Naf, K+, and Ca2+ ions, in the mechanism of GnRH action on superfused R. pipiens hemipituitaries.

MATERIALS AND METHODS Animals

Adult male Rana pipiens (21.3-60 g) obtained from a commercial supplier in Ver-

Address reprint requests to Dr. Paul Licht, Department of Zoology, University of California, Berkeley, CA 94720.

0 1986 ALAN R. LISS, INC.

380 D.A. PORTER AND P. LICHT

mont (G.W. Hazen, Inc.) were kept in holding tanks with running water at 24°C and 12L:12D photoperiod for several weeks to months before use. They were fed crickets.

Superfusion system The superfusion system is detailed in Por-

ter and Licht (‘85). In brief, superfusion columns with an effective volume of 100 pl were fashioned from 1-ml disposable plastic sy- ringes. Pituitaries were removed and bi- sected from frogs subsequent to decapitation, and within 10 min individual hemipituitaries were placed in columns that were at- tached to a Gilson peristaltic pump with Tygon tubing that delivered diluted Dulbec- CO’S modified Eagle’s medium (DME; Gibco) at 25°C at a rate of 4-6 ml/h. The formula- tion of the DME varied according to the requirements of each specific experiment, but always contained 10 mM HEPES (N-2hy- droxyethyl piperazine-N-D-ethanesulfonic acid) and 0.1% bovine serum albumin. When CaClz was omitted from the medium or extra KC1 was added, the concentration of NaCl was adjusted to retain isotonicity. In this study, 20-min fractions of superfusate were collected for subsequent determination of LH and FSH by radioimmunoassay (FtIA). How- ever, since our interests were on long-term responses and to facilitate statistical anal- yses, hourly mean secretion rates were com- puted from these samples; this procedure also minimized variability owing to errors in the timing of sample changes. Hemipituitaries were removed from the superfusion columns at the end of superfusion and homogenized with a glass homogenizer in 500 pl saline for analysis of protein content. Because of high interindividual variability previously observed in these frogs, the basic experimental protocol used throughout this study involved the use of paired hemipituitaries for comparison of two treatments in each series. The paired hemipituitary design did not per- mit the use of a control in normal DME in each case. However, experience with such controls (n = 6) indicates that secretion rates can be expected to remain essentially constant for at least 12 h in DME (Porter and Licht, ’85).

Reagents and hormones The synthetic mammalian GnRH used was

identical with that employed in previous studies in this laboratory, and was provided by J. Rivier (The Salk Institute, La Jolla,

CAI. For all experiments in which GnRH was employed, a dose of 100 nglml was chosen, because this relatively high dose reliably in- duces near maximal secretion of LH and FSH in a biphasic manner (Porter, ’85; Porter and Licht, ’85). Bradford protein assay dye was from BioRad. All other chemicals were from Sigma.

The role of K + The involvement of K + was examined by

two distinct procedures: the use of elevated (lox normal) K+ in the incubation medium and the addition of tetraethyl ammonium chloride (TEA), an agent that should reduce cellular K+ efflux independently of elevated extracellular K’. The effects of each were tested alone and in combination with GnRH- stimulated GtH release.

Effects of 10 x normal (36.3 mM) KC1 All hemipituitaries from five frogs were

initially superfused for 3 h with normal DME. For the following 7 h, one hemipituitary from each pair was superfused with GnRH, while the other hemipituitary was superfused with 36.3 mM KC1.

Effects of high KC1 fr: extracellular Ca2 +

The second series of tests examined the Ca2+ dependence of high KCl effects. Follow- ing 3 h of superfusion with DME, hemipituitaries from two frogs were superfused with medium containing 36.3 mM KC1 with or without CaC12+ €or 7 h. Ethyleneglycol-bi@- amino-ethyl ether)N,N’-tetraacetic acid (EGTA, 0.5 mM) was added to the low calcium medium. Hemipituitaries that had received normal Ca2+ medium were then given Ca2+-depleted medium, and vice versa for the final 2 h of superfusion.

Effects of TEA The first tests examined the effects of TEA

in combination with GnRH. One hemipituitary from each of four frogs was superfused with DME for 3 h followed by 8 h with medium containing GnRH, then for 2 h with DME alone. The paired hemipituitaries were superfused with normal DME for 2 h, then for 1 h with 50 mM TEA. For the next 8 h, these hemipituitaries received GnRH + 50 mM TEA. In the final 2 h of superfusion, the GnRH was removed, but TEA was retained. A second series of tests examined the effects of TEA given in sequence with GnRH. After an initial superfusion with normal DME for

IONS AND FROG GONADOTROPIN SECRETION 381

3 h, hemipituitaries from four frogs then received GnRH for 7 h, followed by 3 h of GnRH + 50 mM TEA, while the paired hemipituitaries were superfused for 7 h with 50 mM TEA, followed by 3 h with TEA + GnRH.

Effects of high KC1 given in sequence with GnRH

Hemipituitaries from four frogs were initially superfused for 3 h with DME. Then, one hemipituitary from each frog was superfused with 36.3 mM KCl for 5 h, followed by 5 h with 36.3 mM KC1 + GnRH. The paired hemipituitaries were superfused for 5 h with GnRH, followed by 5 h with GnRH + 36.3 mM KCI. As a further test of the role of Ca2+ in the action of KC1 (and of or GnRH as well), the hemipituitaries that had received GnRH followed by 36.3 mM KCl were superfused for 5 more h with 36.3 mM KC1 received GnRH followed by 36.3 mM KC1 and GnRH in Ca2+-depleted medium.

The role of Ca2' The role of Ca2+ in the mediation of GnRH

stimulated GtH release was tested by both the removal of Ca2+ from the medium and the addition of a Ca2+ competitor CoC12. The former was similar to tests already described for high K+ actions.

Calcium dependence of GnRH actions One hemipituitary from each of five frogs

was superfused for 3 h with DME, followed by 7 h of superfusion with GnRH. A final 2- h su erfusion with GnRH in the absence of Ca2'(with 0.5 mM EGTA added) followed. The other hemipituitary from each frog was given an initial 3-h superfusion with DME (+ EGTA), then 7 h of superfusion with GnRH in the Ca2+-depleted medium. A final 2 h superfusion with normal Ca2+ (1.2 mM) medium + GnRH was given.

Effects of the calcium competitor CoCl2 One hemipituitary from each of four frogs

was initially superfused for 3 h with DME, followed by 7 h with GnRH. The other hemipituitary from each frog was superfused for 3 h with DME + 2.5 mM CoC12, then for 7 h with GnRH + CoC12.

The role of Nu+

tetrodotoxin (TTX) to prevent Na + influx and replacement of Na+ by choline chloride.

Effects of TTX on GnRH action One hemipituitary from each of four frogs

was superfused for 3 h with DME, followed by 7 h of superfusion with GnRH. For the final 2 h of superfusion, 1 pM of TTX was added to the GnRH-containing medium. The paired hemipituitaries were superfused for 2.33 h with DME, then for 0.67 hour with 1 pM TTX. For the following 7 h, these hemipituitaries were superfused with GnRH + 1 pM W X . In the final 2 h of superfusion, GnRH without ?TX was given.

Effects of Na+ replacement on GnRH action Hemipituitaries from four frogs were su-

perfused for 3 h with DME and then for 7 h with GnRH. The paired hemipituitaries were superfused with DME containing choline chloride in place of NaCl for 3 h before exposure to GnRH in the same medium.

Requirements for Na+ for high K + effects All hemipituitaries from four frogs were

superfused initially with DME for 3 h. Then one hemipituitary from each frog received 7 h of superfusion with 36.3 mM KC1. The other hemipituitary from each animal received 7 h of superfusion with 36.3 mM KC1 +1 pMTTX.

Assays LH and FSH levels in superfusates were

determined with RIA'S developed for R. catesbeiana gonadotropins and validated for R. pipiens (Farmer et al., '77). GtH potencies were determined with a log-logit computer program. Protein content of pituitary homog- enates was determined with the Bradford protein assay using BSA as standard.

Statistics Significant differences between treatment

groups were determined by multivariate analysis of variance (MANOVA) for repeated measures (see Winer, '71) using the micro- computer program SYSTAT (Systat, Inc., Le- land Wilkinson, Evanston, n). The program calculates overall multivariate F-statistics and univariate F-statistics for comparisons between adiacent suDerfusate fractions.

Two methods were em loyed to examine Therefore, sibificant treatment and time ef- the importance of the Na ion in GnRH and fects as well as treatment by time interac- high K+ stimulated GtH release: addition of tions can be readily determined.

P


TABLE 1. Summary of the involvement of ions ( C g + , K+, and Na+) on unstimulated and GnRH-stimulated gonadotropin secretion by superfused frog pituitaries

Treatmenta Figureb Response in gonadotropin secretion'

GnRH 1,3-9 Rapid sustained increase, plateauing after

Involvement of K+ ca. l h

l o x KC1 GnRH + 10xKCl 5 Enhanced GnRH response TEA 4 Relatively rapid increase (slower than

GnRH + TEA 3 74 Enhanced GnRH response

GnRH - Ca2+ 6 No response until Ca2+ added GnRH + CoC1%+ 7 No response l o x KC1 - Ca 2 No response until Ca2+ added

GnRH - Na' 9 Normal GnRH response GnRH + mX 8 Normal GnRH response l o x KCl + "X 10 Normal 10 x KC1 response

1,2,5,10 Gradual progressive increase up to ca. 4 h

GnRH)

Involvement of Ca2+

Involvement of Na+

T3ee text for details of time and concentrations. bRefers to the figures depicting data dealing with each treatment. %sponses refer to patterns of increase over pretreatment levels in hourly secretory rates of both LH and FSH (see figures). Statistical treatments are detailed in the text.

3 4 5 6 7 8 9 10

HOURS OF SUPERFUSION Fig. 1. Mean hourly secretion (+ SEM) of LH and FSH (ng per pg pituitary protein) from

GnRH (X) and 10 x KC1 (squares)-treated hemipituitaries from five frogs. Continuous treatment with either secretagogue began after hour 3.

RESULTS LWFSH ratios were relatively constant both within and among treatments (e.g., mean ra-

Table 1 summarizes the general effects of tios averaged 2.18 + 0.23 among experi- the various ionic and pharmacological treatments). The high degree of correlation be- ments on the secretion of the two gonadotro- tween the two hormones over time is clearly pins. In all experiments, the level of LH evident from the secretion patterns for each secretion was higher than that for FSH, but test (see Figs. 1-10).


I

4.8- FSH

3.6 -

W 2.4- I-

1 a I

z 0 1 I I I I I I I I I I 1 0 - c w Qf 0 W v)

I 18.04 LH I

I

I I I I I I I 1 1 3 4 5 6 7 8 9 10 1 1 12

HOURS OF SUPERFUSION Fig. 2. Mean hourly secretion (+ SEMI of LH and FSH (ng per pg pituitary protein) from

hemipituitaries from two frogs in response to superfusion with l o x KCl for 9 h in the presence (x) or absence (squares) of Ca2+. At hour 10 (shown by vertical dashed line), Ca2+ was added to the previously CaZf depleted medium, and vice versa.

I I I

t

I I I I I

48.0

36.0

24.0

12.0

0 3 4 5 6 7 8 9 1 0 1 1 1 2 1 3

L

, 8 9 10

HOURS OF SUPERFUSION Fig. 3. Mean hourly secretion (+ SEM) of LH and FSH (ng per pg pituitary protein) by

hemipituitaries from four frogs in response to superfusion with GnRH (X) or with GnRH + 50 mM TEA for hours 3-11. Normal DME (x) or DME + TEA (squares) was given in the final 2 h (shown by vertical dashed line).



hemipituitaries from four frogs superfused with GnRH for 7 h, followed by 3 h with GnRH + 50 mM TEA (x), or superfused with 50 mM TEA for 7 h, followed by 3 h with TEA + GnRH (squares); time of change in medium composition shown by vertical dashed line.

21.0 FSH

w '5'75 10.5 1 I I A

I I

"I I I I I I I I I I I 3 4 5 6 7 8 9 10 11 12 13


hemipituitaries from four frogs superfused for 5 h with lox KC1, then 5 h with lox KCl + GnRH (x), or for 5 h with GnRH followed by 5 h of superfusion with GnRH + l o x KCI (squares); time of change in medium composition shown by vertical dashed line.

IONS AND FROG GONADOTROPIN SECRETION

I2.O-

385

FSH

10.8 -

w 7.2- c a

= o I I I I I I

E 3.6-

I

0 I I I 1

HOURS OF SUPERFUSION Fig. 6. Mean hourly secretion (+ SEMI of LH and FSH (ng per pg pituitary protein) by

hemipituitaries from four frogs in response to superfusion with GnRH in the presence ( x ) or absence (squares) of Ca2+ for 7 h (X I . Ca2+ was added to the medium in the final 2 h of superfusion (shown by vertical dashed line) where it had been previously absent, and vice versa.

Fig. 7. Mean hourly secretion (+ SEM) of LH and FSH (ng per pg pituitary protein) by hemipituitaries from four frogs in response to continuous superfusion with GnRH for 7 h (XI, or to superfusion with GnRH in the presence of 1.2 mM CoClz (squares).


24.0- FSH

18.0-

w 12.0- I-

E 6.0- a

I

3 4 5 6 7 8 9 10 1 1 12

HOURS OF SUPERFUSION

Fig. 8. Mean hourly secretion (k SEM) of LH and FSH (ng per pg pituitary protein) by hemipituitaries from four frogs in response to superfusion with GnRH in the presence (squares) or absence (x) of tetrodotoxin for 7 h. In the final 2 h (shown by vertical dashed line), 'M'X was added to the medium where it has been previously absent (X), and vice versa.

18.0- FSH

13.5-

9.0 - w I- Q: 4.5- w z o I I I I I I I

- 1 I I I I I I I

3 4 5 6 7 8 9 10

HOURS OF SUPERFUSION

Fig. 9. Mean hourly secretion (* SEM) of LH and FSH (ng per pg pituitary protein) by hemipituitaries from four frogs in response t o continuous superfusion with GnRH for 7 h in DME (x) or with GnRH in medium in which NaCl had been replaced with choline chloride (squares).

IONS AND FROG GONADOTROPIN SECRETION

0

387

I I I I I I I

W l-

oi a

Fig. 10. Mean hourly secretion (& SEM) of LH and FSH (ng per pg pituitary protein) by hemipituitaries from four frogs in response to 7 h of continuous superfusion with 10 x KCI in the presence (squares) or absence ( X ) of tetrodotoxin.

Although absolute levels of gonadotropin secretion varied several-fold among experiments, even for the same treatment, the standard dose of GnRH used in most tests (Figs. 1, 3-9) produced a consistent pattern of response: LH and FSH output increased significantly within the first 20 min sample and tended to plateau after 1 h, remaining relatively constant throughout the several additional hours of superfusion. This response serves as useful reference for other treatments.

The role of K+ Hemipituitaries were stimulated by 36.3

mM KCI, but GtH secretion increased more gradually than with GnRH and did not peak until approximately 4 h. In contrast to GnRH, neither FSH nor LH output increased significantly during the first hour of high KC1 treatment. By hour 7, GtH levels were higher (significantly so for LH, P < .03) then at the end of 1 h of superfusion with high KC1, and GtH levels plateaued thereafter. GtH levels in response to high KC1 were significantly lower than in response to GnRH through approximately 4 h of treatment (P < .001 for LH, P < .008 for FSH), beyond which differences were insignificant, although continuing to average lower with

high Kf than with GnRH. The same temporal pattern of response to elevated K+ was also consistently observed in subsequent tests (cf. Figs. 2,5, 10).

Exclusion of CaClz from the medium significantly attenuated LH secretion and fully blocked FSH secretion in the presence of high KC1 (Fig. 2). Also, when Ca2+ was subse- quently removed from the hemipituitaries that had received K+, GtH levels declined after 2 h. In contrast, secretion of FSH and LH increased when Ca2' was restored for the final 2 h to hemipituitaries that had been su erfused with high K+ in the absence of C$+. The importance of Ca2+ in the action of high K+ is further indicated by the results shown in Table 2. The removal of Ca2+ from the medium containing 100 ng/ml GnRH and 36.3 mM KC1 led to a rapid decline in LH and FSH secretion within 2 h; a significant reduction occurred in LH by 3 h, and in FSH levels by 2 h after Ca2+ removal.

Secretion of both gonadotropins was significantly higher when GnRH was given in combination with TEA then with GnRH alone for the first 3 h (P < .005 for LH, P < .009 for FSH, Fig. 3). GtH levels remained higher for pituitaries superfused with GnRH + TEA thereafter, but differences were not significant. GtH levels declined for both


TABLE 2. Effect of calcium removal subsequent to superfusion with DME + 36.3 mM KC1 -t 100 ng/ml GnRH on meana LH and FSH secretion from four Rana pipiens

hemiDituitaries

Significanceb Time LH FSH LH FSH

Start' 7.1 k 1.2 4.5 f 1.0 - - Hour

1 6.7 k 1.5 3.3 * 1.2 < .65 < .2 2 3.0 f 0.6 1.8 + 0.8 < .1 < ,002 3 2.1 k 0.6 1.0 + 0.5 < .006 < .01 4 1.4 k 0.2 0.8 f 0.3 < .02 < .01 5 1.6 + 0.21 0.6 + 0.2 < .03 < .02

%g/pg pituitary proteidfraction. bFirst value is for difference between LH levels, second value is for FSH contrasts for each hour vs the start. 'GtH values after 5 h of superfusion with GnRH f36.3mM KCl, subsequent times indicate number of hours of superfusion in Caz+-depleted medium with GnRH + 36.3 mM KC1.

treatments to near pretreatment values subsequent to removal of GnRH.

Secretion of LH increased significantly from hemipituitaries superfused with either 100 ng/ml GnRH or 50 mM TEA after the 3- h rinse (P < .005; Fig. 4). There were also significant increases in LH levels for both treatments between hours 6 and 8 (P < .026). The addition of GnRH to the hemipituitaries that had been receiving TEA caused a significant rise in LH secretion, as did addition of TEA to the tissues that had been receiving GnRH Cp < .048). GtH increased at a significantly faster rate for tissues first superfused with TEA followed by addition of GnRH than for tissues treated with the two in the reverse order (P < .021). The combination of GnRH and high K+ led to significantly higher secretion of both gonadotropins (Fig. 5: P < .0001 for LH, P < .017 for FSH) than that which was due to the presence of either secretagogue alone.

The role of Ca2' When GnRH was superfused in Ca2+-de-

pleted DME, neither gonadotropin was se- creted at levels significantly above pre-GnRH treatment values (Fig. 6). Superfusion of these hemipituitaries with DME containing CaC12 + GnRH for the final 2 h caused a significant rise in LH (P < .004) and FSH (P < .032) secretion. Superfusion with GnRH in CaCl2-depleted medium for the final 2 h caused secretion of both gonadotropins from the paired hemipituitaries to decline slightly, but not to pretreatment levels.

Superfusion of hemipituitaries with GnRH + 2.5 mM CoC12 resulted in significantly less LH (P < .002) and FSH (P < .007) secre-

tion than did GnRH in normal medium (Fig. 7). Pretreatment GtH levels were also some- what reduced from hemipituitaries superfused with DME + CoC12.

The role of Nui The presence of 1 pM TTX had no signifi-

cant effect on levels of basal or GnRH stimulated GtH secretion (Fig. 8). Replacement of NaCl with choline chloride also had no significant effect on responsiveness to GnRH (Fig. 9). Pretreatment levels of LH and FSH were unaffected by the exclusion of NaC1. ?yrX had no significant effect upon FSH or LH secretion in response to high K+ (Fig. 10).

DISCUSSION

These studies demonstrate a differential dependence of pituitary gonadotropin secretion on the three cations tested. In general, Ca2+ appears to be most important, while Na+ has the least effect on the response to GnRH. K+ may affect the response to GnRH, but its specific involvement in GnRH activity remains unclear (Table 1). The conspicuous parallelism between the secretory patterns and consistent ratios of FSH and LH throughout all treatments indicates that the ionic dependencies of secretion are similar for the two gonadotropins.

Treatment of pituitary tissue with elevated KC1 (between 20 and 60 mM) has been reported to cause increased secretion of gonadotropins in several mammalian species, although durations and magnitudes of responses have been variable (for example, rats-de Koning et al., '78; Bourne and Bald- win, '80; Smith, '85; cows-Zolman and Val-


enta, '81; sheep-Adams and Nett, '79). High Kf also stimulates the release of LH from chicken anterior pituitary cells (Bonney and Cunningham, '77). Thornton and Geschwind ('74) showed that 28 mM KC1 caused a pro- nounced increase in LH secretion from R. pipiens pituitary pieces, but the time course of this ion effect was not defined in their static incubations. The secretory profile in response to superfusion with elevated K+ displayed by the frog hemipituitaries in the present study closely resembles that shown by the superfused rat pituitaries described by Bourne and Baldwin ('80), but Smith ('85) reported a more rapid response to high Kf in glands from lactating rats. In general, there do not appear to be consistent differences among the tetrapod vertebrate pituitaries: High K+ appears to be more effective in stimulating the delayed (phase 2) release of GtH than the acute (phase 1) release.

Several studies have indicated that Ca2+ must be present in the medium for elevated Kf levels to be effective in stimulating pituitary hormone release from mammalian (Ad- ams and Nett, '79; Bourne and Baldwin,'80; Kame1 and Krey, '83; Kraicer et al., '69; Zol- man and Valenta, '81; Wakabayashi et al., '69; Samli and Geschwind, '68) and avian pituitary tissues (Bonney and Cunningham, '77). Moreover, other studies with mammalian tissues have shown that elevated Kf levels may act by increasing Ca2+ flux (for review, see Trifaro, '77; also, Bicknell and Schofield, '77; Aizawa and Hinkle, '85; Tar- askevich and Douglas, '78; Tashjian et al., '78; Williams, '76). The greatly attenuated reponses to high K+ in Ca2+-depleted medium by the frog pituitaries (Table 2, Fig. 2) are consistent with the results in other species; as in mammals, high K+ may act by stimulating Ca2+ influx.

Taraskevich and Douglas ('78) showed that catecholamines (which can suppress prolactin secretion from fish-Alosa pseudoharen- gus-pituitary cells) function by suppressing action potentials that possess both Na+ and Ca2' components. The Ca2+ component was identified by blocking K + conductance (that presumably acts by increasing Ca2+ flux- see above) with 10 mM TEA in medium that also contained TTX. The TEA caused large regenerative potentials that were sup ressed by the replacement of Ca2+ with Mn'+. For the present study, a dose of 50 mM TEA was chosen because of its effectiveness in main- taining depolarization of neural tissue (Her-

man and Gorman, '81). This dose of TEA was comparable to 100 ng/ml GnRH in stimulating increased GtH secretion. These results suggest that, as in the mammal, elevated K+ results in depolarization (which can be mim- icked by preventing repolarization with the addition of TEA), causing Ca2+ flux and thus elevated GtH secretion.

The significance of the results with TEA lies in the indication that it is the depolarization, not the K+ flux per se, that causes GtH secretion to rise. Interestingly, TEA caused more rapid increases in GtH secretion than did high K'. The apparent additivity of the effects of TEA and GnRH (Figs. 3,4) suggests that high Kf and GnRH, as in some mammalian studies (Zolman and Valenta, '81; Wakabayashi et al., '69, '73; Samli and Geschwind, '68), may not act in precisely the same manner to stimulate GtH secretion. The different temporal secretory profiles elic- ited by GnRH and high K+ also indicate divergent mechanisms: GnRH when given alone or added to high K+ Fig. 5) caused a more rapid increase in GtH secretion than did comparable treatment with high K+ . Al- though the effects of high K' and GnRH are apparently additive, Ca2+ appears essential for the mechanism of action of both.

Although high Kt mimics some of the actions of GnRH, several lines of evidence suggest that GnRH may not stimulate GtH secretion by altering K+ fluxes. The normal role of K+ flux may be to promote repolarization subsequent to the depolarizing action of GnRH, thereby dampening the response to GnRH. TEA + GnRH produces an exagger- ated version of the profile owing to GnRH alone, whereas in superfusions with high K+, the profiles are always clearly different from GnRH alone. Thus, TEA may not act in precisely the same manner as does high K+. Further experiments with K+-depleted media or with specific K+ ionophores would be useful.

Calcium has been implicated in the action of GnRH in numerous studies employing mammalian tissues. These studies have shown that the absence of Ca2+ strongly attenuated the responsiveness to GnRH (Conn and Rogers, '80; Adams and Nett, '79; Bourne and Baldwin, '80; Zolman and Valenta, '81; Wakabayashi et al. '69). Similar results were obtained in studies with chicken pituitary cells (Bonney and Cunningham, '77; Luck and Scanes, '80). The relative importance of Ca2+ for phase 1 versus phase 2 secretion


may differ. For example, Bourne and Bald- win ('80) reported that in the rat, phase 1 secretion was completely inhibited by the lack of Ca2+, whereas phase 2 secretion was only partially inhibited. Inhibition may be more complete in the frog (Fig. 6): There was only an insignificant increase in GtH secretion in the latter half of the 7 h superfusion in the absence of Ca2+. The sharp increase in GtH secretion following the replacement of Ca2+ to the depleted medium after 10 h indicates that the secretory mechanism had not been damaged by the prolonged absence of this ion.

A subject of contention for the mammalian gland is whether Ca2+ must be available intracellularly or extracellularly. Conn and Rogers ('80) and Bates and Conn ('84) found that D600 (an agent that blocks Ca2+ channels) impeded the effectiveness of GnRH, in- dicating the importance of extracellular Ca2'. Biales et al. ('77) found that 7.5 mM C0C12 could be used to block the Ca2' component of action potentials in GH3 pituitary cells; thus, this competitor should provide evidence for the dependence on extracellular Ca2'. Accordingly, the conspicuous attenua- tion of the responsiveness to GnRH in the presence of 2.5 mM CoCl2 (Fig. 7) indicates that Ca2+ influx is also important for GnRH action in the frog.

Other studies have sug ested the importance of intracellular Ca'+ in the mechanism of GnRH action. For example, rat cells prelabeled with 45Ca2+ released Ca2+ to the extracellular environment before or simulta- neous with LH release (Williams, '76). The relatively slow decline in GtH secretion by frog glands upon removal of Ca2+ from the medium suggests that intracellular Ca2' pools may be sufficient to support GtH secretion for some time. Similar results were observed upon Ca2+ removal in superfusions with high K+ (Fig. 2) and in superfusions with high K+ + GnRH (Table 2). Thus, it is likely that both intracellular and extracellular Ca2+ are required for normal action of GnRH in the frog.

Sodium has been implicated as a significant ion in the release of some pituitary hormones. Conn and Rogers ('80) found that veratridine (which allows passive Naf influx) stimulated the release of LH from rat pituitary cell cultures. This effect was blocked by lop5 M ?TX, but the action of

GnRH was not affected by ?TX. Tl'X (2 x lop6 M) eliminated spontaneous action potentials in prolactin cells from alewifes (Tar- askevich and Douglas, '78). In the present study, prevention of Naf influx through TTX-sensitive channels was without significant effect on the action of GnRH or high Kf (Fig. 10). Replacement of NaCl with choline chloride also confirms that Na + is not essential for the action of GnRH (Fig. 9).

One impetus for this study was to deter- mine if the difference in resistance to GnRH desensitization between mammals and frogs might be due to mechanistic differences in GnRH action, in particular, in the roles of K+, Ca2+, and Na+. However, the physiol- ogy of these ions does not appear to underlie the observed difference in GnRH responsiveness between frogs and mammals. The re- quirement for Naf in GnRH action is minimal at best. The diffusion of K+ out of the gonadotroph after GnRH stimulation, which reduces the state of depolarization, may in fact define a physiological role for this ion. The presence of Ca2+ is critical, as its absence from the medium eliminates phase 1 secretion and greatly attenuates phase 2 secretion of GtH in response to GnRH. This involvement in both phases of the GnRH response suggests that Ca2' may act at more than one site. In a review by Trifaro ('77), a model was described in which Ca2+ is thought to act as a bridge between anionic groups at the plasma membrane and the membrane of the secretory granule, thus facilitating exocytosis. This model may ex- plain in part the involvement of Ca2+ in the very rapid initial phase of GtH secretion in response to GnRH, but probably does not account for much of the action of Ca2' in phase 2 GtH secretion. The overall similarity in the roles of these ions in frogs and mammals, as well as other tetrapods, suggests evolution- ary conservatism in the mechanism of GnRH action in regulating pituitary GtH secretion.

ACKNOWLEDGMENTS

We thank Raymond Pang for technical as- sistance. This work was supported by NSF grant PCM 8406751.

LITERATURE CITED

Adams, T.E., and T.M. Nett (1979) Interaction of GnRH with anterior pituitary. III. Role of divalent cations, microtubules and microfilaments in the GnRH-acti- vated gonadotroph. Biol. Reprod., 21:1073-1086.


Aizawa, T., and P.M. Hinkle (1985) Differential effects of thyrotropin-releasing hormone, vasoactive intestinal peptide, phorbol ester, and depolarization in GH& rat pituitary cells. Endocrinology, 116:909-919.

Bates, M.D., and P.M. Conn (1984) Calcium mobilization in the pituitary gonadotrope: relative roles of intra- and extracellular stores. Endocrinology, 115:1380-1385.

Biales, B., M.A. Dichter, and A. Tischler (1977) Sodium and calcium action potentials in pituitary cells. Na- ture, 267~172-174.

Bicknell, R.J., and J.G. Schofield (1977) Mechanism of action of somatostatin: Growth-hormone release and 45-Ca calcium ion translocation caused by increasing the extracellular potassium ion concentration. Biochem. Sot. Trans., 5:1051-1053.

Bonney, R.C., and F.J. Cunningham (1977) Effect of ionic environment on the release of LH from chicken anterior pituitary cells. Mol. Cell. Endocrinol. 7:245-251.

Bourne, G.A., and D.M. Baldwin (1980) Extracellular Ca-independent and -dependent components of the biphasic release of LH in response to luteinizing hormone-releasing hormone in vitro. Endocrinology,

Conn, P.M., and D.C. Rogers (1980) Gonadotropin release from pituitary cultures following activation of exoge- nous ion channels. Endocrinology, 107:2133-2134.

deKoning, J., J.A.M.J. van Dieten, and G.P. Rees (1978) Refractoriness of the pituitary gland after continuous exposure to luteinizing hormone releasing hormone. J . Endocrinol., 79:311-318.

Farmer, S.W., P. Licht, H. Papkoff, and E. Daniels (1977) Purification of gonadotropins in the leopard frog (Rana pipiens). Gen. Comp. Endocrinol., 32:158-162.

Herman, A., and A.L.F. Gorman (1981) Effects of tetra- ethylammonium on potassium currents in a mulluscan neuron. J. Gen. Physiol., 78:87-110.

Kamel, F., and L.C. Krey (1983) Gonadal steroid modu- lation of LH secretion stimulated by LHRH, Ca2+ and CAMP. Mol. Cell. Endocrinol., 32285-300.

Kraicer, J., and J.V. Milligan (1971) Effect of colchicine on in vitro ACTH release induced by high K + and by hypothalamic-stalk median eminence extract. Endocri- nology, 89:408-412.

Kraicer, J., J.V. Milligan, J.L. Gosbee, R.G. Conrad, and C.M. Branson (1969) In vitro release of ACTH: Effects of potassium, calcium and corticosterone. Endocrinol- ogy, 85: 1144-1153.

Liu, T.C., and G.L. Jackson (1978) Modifications of luteinizing hormone biosynthesis and release by gonadotropin-releasing hormone, cycloheximide, and actinomycin D. Endocrinology, 103:1253-1263.

Luck, M.R., and C.G. Scanes (1980) Ionic and endocrine factors influencing the secretion of luteinizing hormone by chicken anterior pituitary cells in uitro. Gen. Comp. Endocrinol., 41:260-265.

McCreery, B.R., and P. Licht (1983a) Pituitary and gonadal responses to continuous infusion of gonadotropin releasing hormone in the male bullfrog, R ~ R U catesbeiana. Biol. Reprod., 29:129-136.

107:780-788.

McCreery, B.R., and P. Licht (1983b) Induced ovulation and changes in pituitary responsiveness to continuous infusion of gonadotropin-releasing hormone (GnRH) during the ovarian cycle in the bullfrog, Rana catesbeiana. Biol. Reprod., 292363471.

Pickering, A.J.M.C., and G. Fink (1976) Priming effect of luteinizing hormone releasing factor: In uitro and in uiuo evidence consistent with its dependence upon protein and RNA synthesis. J. Endocrinol., 69:373-379.

Porter, D.A. (1985) The effects of cycloheximide on In vitro response of Rana pipiens pituitaries to continuously superfused gonadotropin-releasing hormone (GnRH). Biol. Reprod., 33t393-400.

Porter, D.A., and P. Licht (1985) Pituitary responsiveness to superfused GnRH in two species of Ranid frogs. Gen. Comp. Endocrinol., 59:308-315.

Samli, M.H., and 1.1. Geschwind (1968) Some effects of energy-transfer inhibitors and of Ca2+-free or K+-enhanced media on the release of luteinizing hormone (LH) from the rat pituitary gland in vitro. Endocrinol- ogy, 82:225-231.

Smith, M.S. (1985) Release of luteinizing hormone and follicle-stimulating hormone from pitui'taries of lactating rats by gonadotropin-releasing hormone and high potassium concentration. Endocrinology, 116:1826- 1834.

Taraskevich, P.S., and W.W. Douglas (1978) Catechol- amines of supposed inhibitory hypophysiotrophic function suppress action potentials in prolactin cells. Nature, 2769332-834.

Tashjian, Jr., A.H., M.E. Lomedico, and D. Maina (1978) Role of calcium in the thyrotropin-releasing hormone- stimulated release of prolactin from pituitary cells in cuIture. Biochem. Biophys. Res. Commun. 8lr798-806.

Thornton, V.F., and 1.1. Geschwind (1974) Hypothalamic control of gonadotropin release in amphibia: Evidence from studies of gonadotropin release in vitro and in vivo. Gen. Comp. Endocrinol., 23:294-301.

Trifaro, J.M. (1977) Common mechanisms of hormone secretion. Ann. Rev. Pharmacol. Toxicol., 17:27-47.

Wakabayashi, K., I.A. Kamberi, and S.M. McCann (1969) In vitro response of the rat pituitary to gonadotropin- releasing factor and to ions. Endocrinology, 85t1046- 1056.

Wakabayashi, K., Y. Date, and B.4. Tamaoka (1973) On the mechanism of action of luteinizing hormone-releasing factor and prolactin release inhibiting factor. En- docrinology, 923398-704.

Williams, J.A. (1976) Stimulation of 45Ca++ efflux from rat pituitary by luteinizing hormone-releasing hormone and other pituitary stimulants. J. Physiol. (Land.), 260: 105-115.

Winer, B.J. (1971) Statistical Principles in Experimental Design. 2nd. ed. McGraw-Hill: New York.

Zolman, J.C., and L.J. Valenta (1981) Effect of calcium and potassium on basal and gonadotropin-releasing hormone (GnRH) stimulated release of luteinizing hormone &HI. Acta. Endocrinol. 9726-32.

Dependence of GnRH action on Na+, K+, and Ca2+ in the frog, Rana pipiens, pituitary

Documents

Transcript of Dependence of GnRH action on Na+, K+, and Ca2+ in the frog, Rana pipiens, pituitary