Molecular and Cellular Endocrinology, 76 (1991) 63-70 0 1991 Elsevier Scientific Publishers Ireland, Ltd. 0303-7207/91/$03.50

63

MOLCEL 02456

The effects of age on the postreceptor regulation of luteinizing hormone secretion by gonadotropin-releasing hormone Serge B&sle ‘, Diego Bellabarba 2, Jean-Guy Lehoux 3 and Nicole Gallo-Payet * ’ Department of Obstetrics-Gynecology, Faculty of Medicine, University of Montreal, Montreal, Quebec, Canada, and Departments of ’ Medicine and 3 Biochemistry, Faculty of Medicine, University of Sherbrooke, Sherbrooke, Quebec, Canada (Received 14 September 1990; accepted 4 December 1990)

Key words: Aging; Luteinizing hormone; Gonadotropin-releasing hormone; Phosphoinositide; Protein kinase C

We studied the effects of age on the roles of phosphoinositide (PI) and protein kinase C (PKC) in luteinizing hormone (LH) release by gonadotropin-releasing hormone from mouse pituitaries. Pituitary cells from intact and 14-day ovariectomized (OVX) mice aged 4-8 months, lo-12 months and 14-18 months were cultured at a dilution of 3 X lo5 cells/ml of Nil99-bovine serum albumin medium for 3 days prior to stimulation with either buserelin or phorbol ester (phorbol my&ate acetate, PMA), while LH was assayed by radioimmunoassay using anti-rat LH antibody (NIDDK-5-10). In intact young mice, buserelin and PMA specifically induced time- and dose-dependent increases in LH release with specific mean EDso of 0.82 x 10’ I1 M (buserelin) and of 1.6 x lo-* M (PMA) and a maximal LH release of 138 f 15 ng/106 cells after a 3 h stimulation period. Age did not affect the EDso of either agonist but significantly reduced their ability to release LH. This reduction was more pronounced for buserelin than for PMA and was evident as early as middle-age. OVX resulted in a significant increase in both basal and stimulated LH release, but did not affect the age-related reduced secretion rate of LH by either agonist. Buserelin stimulated the incorporation of [3H]inositol into [3H]inositol phosphates (IP) in a dose-dependent manner, which was unaffected by either age or OVX. We conclude that, with aging, there occurs a reduced LH release rate to both buserelin and PKC stimulations, uncoupled to changes in PI-IP cycle. Furthermore our results suggest that these cellular defects are first manifested at middle-age and are unaffected by the status of the gonad.

Introduction Aging in the female rodent is associated with gradual reductions in gonadotropin secretion and

Addressfor correspondence: SergeB&isle,Departmentof Obstetrics and Gynecology, Faculty of Medicine, University of Montreal, C.P. 6128, Succursale A, Montreal, Quebec H3G 357, Canada. This work was supported by a grant from MRC Canada (MA-6732, S.B.).

reproductive cyclicity. Thus, as reviewed by Steger et al. (1982), aging mice demonstrate luteinizing hormone (LH) deficits before the appearance of ovarian acyclicity and manifested by delays in preovulatory surge, declines in peak amplitude, blunted elevations to castration and attenuated positive and negative feedback responses. In contrast to these observations, the endocrine mechakm(s) responsible for these age-related changes in pituitary gonadotropin production are poorly understood. Some reports have suggested a de-

64

crease in pituitary responsiveness to gonadotropin hormone-releasing hormone (GnRH) as being partly responsible for these accelerated changes in female reproductive functions (Marian et al., 1981). However, the degree of pituitary involvement in these deficiencies is still controversial (Cooper et al., 1984). To clarify this issue, in previous studies we undertook experiments on the interaction of GnRH with its pituitary receptor throughout aging in mice and have observed no significant changes in the up- and down-regulation kinetics of the hormone-receptor complex. However, we did observe age-related changes in pituitary gonadotropin secretion originating from cellular defects distal to the binding of GnRH to its receptor (B61isle et al., 1990)o Recent evidence suggests that the formation of the GnRH-receptor complex activates a phospholipase C which hydrolyzes phosphatidylinosiu tol 4,5-bisphosphate (PIP2), leading to generation of inositol triphosphate (IP3) and diacylglycerol (DAG) (Andrews and Conn, 1986). These breakdown products can act as second messengers within gonadotropes, the IP3 mobiliT~ingintracellular calcium stores while DAG can activate the calcium/phospholipid-dependent protein kinase C (PKC) (Berridge and Irvine, 1984; Nishizuka, 1984). The PKC pathway, in synergy with the calcium-cahnodulin system, will lead to phosphorylation of intracellular proteins and to LH release (Catt and Stojilkovic, 1989). The aim of the present study was to evaluate the role of these cellular messengers in gonadotropin secretion throughout aging. Since phorbol esters can substitute for DAG and activate PKC to stimulate LH secretion (Johnson et al., 1988), we compared the stimulatory effects of these probes with that of a GnRH agonist in gonadotropin secretion. Materials and methods Animals and tissues

Neonatal C57BL/6J female mice were purchased from Jackson Laboratory (Bar Harbor, ME, U.S.A.) and maintained in groups of 4-6 mice per cage in a limited access aging colony. Cycling young (4-8 months) and middle-aged (10-12 months) as well as acyclic old (15-18 months) mice were studied either intact (metes-

trous or constant anestrous, respectively) or 2 weeks after bilateral ovariectomy (OVX). On the day of experimentation, animals were sacrificed by decapitation, the skull was opened azld the pituitary dissected and collected into sterile Hepes-buffered (10 mM) medium 199 containing 0.3% bovine serum albumin (BSA) (199/BSA; Sigma, St. Louis, MO, U.S.A.) and 20 mg/ml gentamicin (Schering Corp., Bloomfield, N J, U.S.A.) at room temperature, Methods

Cell dispersion and culture techniques were performed according to methods pubhshed by Andrews and Corm (1986). The pituitaries were minced in 30 ml 199/BSA and washed by centrifugation twice. Tissue fragments were then digested enzymatically with solution A (199/BSA containing 0.25% coilagenase (126 U/mg; Worthington CLSII, Freehold, N J, U.S.A.), 0.17o hyaluronidase (355 U/mg; Sigma) and 30 pg/ml DNase I (Sigma; 1950 Kunitz units/mg protein)) in a Dubnoff metabolic shaker under agitation at 100 cycles/rain at 37°C. At 5 rain intervals, the tissue suspension was gently passed through a sterile serological pipette 4 or 5 times. After 15 rain of treatment, the suspension was decanted into a second sterile 50 ml centrifugation tube and the residual tissue fragments digested again. The combined cell suspension was brought to a volume of 50 ml with 199/BSA and centrifuged at 200 × g for 10 min. The pellet was then resuspended in 10 ml of 199/BSA solution (B) containing 0.257o collagenase, 0.27o hyaluronidase, 30 /~g/ml DNase and 4 mM EGTA (Sigma). The centrifuge tube was capped, placed on its side in a water bath at 37°C and agitated. At 5 min intervals, the tissue suspension was again passed through a sterile serological pipette 4 or 5 times. After 25 rain, the cell suspension was then decanted through a 400 /~m nylon mesh cloth. 10 ml of sterilized solution B were added to the cell suspension, which was incubated for 15 rain. The filtrate was brought to a volume of 50 ml with 199/BSA containing 4 mM EGTA and cooled at 4°C. After contrifugation at 200 × g for 10 min, the cell pellet was resuspended in culture medium (199/BSA) for further study. A mean yield of 3-5 × 105 cells/

65

pituitary was obtained and 90-95% of the cells were viable based on tr~gan blue dye exclusion. Cell suspension was ~,,:c;ught to a concentration of 3 × 105 cells/1.5 ml 199/BSA containing 10% horse serum, 2.5% fetal calf serum (Whittaker M.A. Bioproducts, Walkerville, MD, U.S.A.) and 20/zg/ml gentamicin in Petri Falcon-3001 dishes. They were maintained at 37 *C for 3 days prior to study, in a controlled atmosphere (95% air, 5% CO2).

Radioimmunoassay of LH Highly purified rat LH (NIDDK-LH-I-7), antiserum to rat LH (NIDDK-anti-rLH-S-10) and reference preparation (NIDDK-rLH-RP-3) were obtained from the National Institute of Diabetes and Digestive and Kidney Diseases directed by Dr. A.F. Parlow. The rLH radioimmunoassay (RIA) was performed as directed by the N I D D K cover instructions, except that [125I]LH was prepared by the lodogen technique, as previousl3 reported (B61isle et al., 1990). Bound and free hormones were separated on Sephadex 2G-100 columns. Pituitary cells were incubated with bioassay medium 199/BSA without or with buserelin ((D-Ser (tBu) 6, Pro9 NHEt) GnRI-I; Hoechst, Montreal, Canada) or phorbol 12-myristate ester (PMA; LC Services, Woburn, MA, U.S.A.) and the intra- and extracellular milieu assayed for LH content. We have previously validated the sensitivity, specificity and precision of this assay (B61isle et al., 1990) for mouse LH determinations. Measurement of inositol phosphate accumulation These determinations were performed accordJng to methods published by Gallo-Payet et al. (1986). After 3 days in culture, the medium was discarded and the cells were washed twice in cold phosphate buffer saline (PBS, NaCI 135 mM, KCI 4.5 mM, CaCl~ 0.4 mM, MgCI2 0.5 mM, NaH2PO4 1.0 mM, pH 7.4) supplemented with 5.6 mM glucose, 0.3% BSA, 10 mM Hepes and 5 mM LiC1 (PBS/LiC1). They were then incubated in PBS/LiCI with 6.7 ~Ci/ml [3H]myo-inositol (1,2N-[3H]inositol; 45-60 Ci/mmol, New England Nuclear, Boston, MA, U.S.A.) for 5 h at 37°C under water-saturated atmosphere. After this labeling period, the cells were washed twice in PBS/LiCI medium and further incubated for 15

min at 37°C. The medium was then removed and the cells were incubated for 15 min at 37°C with 700/tl PBS glucose (1 g/liter) and BSA (1 mg/ml), LiCI (10 mM) and Hepes (10 mM) and various amounts of buserelin (10 -13 to 10 -6 M). The incubation was terminated by rapidly removing the medium and adding 1 ml 5% perchloric acid and 200 pl BSA (20 mg/ml). After a rapid centrifugation of acid extracts, total inositol phosphates (IP) in tile supernatant were :~eparated by chromatography on Dowex 1 x 10 (100-200 mesh; formate form) columns (4.0 x 0.6 cm) as described (Berridge et al., 1983). Labeled inositol iipids were determined after c h l o r o f o r m / m e t h a n o l / H C I (1.00: 100 : 1) extraction, as described (Creba et al., 1983). To discriminate phosphatidylinositol (PI), phosphatidylinositol 4-phosphate (PIP) and PIP2, the extracted inositol lipids were deacylated with NaOH (1 M) in the presence of MeOH, and the resulting products were separated by chromatography as described above. In preliminary experiments we validated our assay by varying the following conditions: cell concentration, doses of buserelin, and time of 3H labeling to reach isotopic equilibrium. Due to the short response time in IP accumulation (5-15 rain) we were unable to detect any significant release of LH in the medium as determined by RIA.

Statistics Two-way analysis of variance was used to evaluate LH release rates under various experimental conditions and for [3H]IP accumulation. Paired Student's t-tests were used to study the effects of age on the ability of the pituitary to secrete LH in response to 10 -7 M buserelin or PMA as well on [3H]IP formation after stimulation with buserelin. Results

LH secretion in vitro Fig. 1A and B summarizes some of the experiments validating our in vitro studies on LH secretion by young mouse pituitary. As shown in Fig. 1A with cells cultured for 3 days prior to study, buserelin (10 -7 M) caused a progressive increase in LH release, with a near-maximum response after 3 h of incubation. This amount represented

66

~, 15o ,~ lOO, g v

=;

so O -1 24

12

0123

mCU~110N11ME(HOUeS)

/.

" ~ 155

g t-

100 o

O

0

-13

~

-12

-11

-~0

-9

Log BUSmEUN(U)

Fig. 1. Time and dose-response curves of the effects of buserelin and PMA on LH release. Cultured pituitary cells from young intact female mice were cultured 72 h prior to stimulatio, studies under conditions as described in Materials and Methods. A: Cells were stimulated with buffer containing no (controls, o . o) or 10 -7 M of either buserelin ( 1 1 ~ 1 1 ) or PMA (e e) for the indicated periods of time. B: Pituitary cells were stimulated for 3 h at 37°C with increasing doses of buserelin without (o o), or with 10 -t° M ( O - - - - - - O ) , 10 -9 M (,, -) or 10 -s M (A &) of PMA. /kfter the stimulation studies, LH was determined in the culture medium using an NIDDK RIA kit. Each data point represents the mean + SE LH release of six se~.,arate experiments, each performed in triplicate (where not shown, SE are smaller than the data point).

Fig. 2 compares the effects of buserelin and PMA on LH release by pituitary cells from young and old intact mice. As can be seen, in young animals both agonists caused dose-dependent increases in LH release over the cop.centration range from 10 -13 to 10 -6 M, with an EDso of 0.82 +_ 0.07 x 1 0 - n M for buserelin and of 1.6 + 0.04 × 10-8 M for PMA. A 1 h preincubation period with 10 -6 M of a bioactive G n R H antagonist (o-Phe 2'6, Pro 3LHRH) completely abolished the effects of buserelin (results not shown) while the PlViA-induced release of LH was maintained. Conversely, preincubation with staurosporin 10 -6 M prevented the stimulatory effects of PMA without affecting those of buserelin (results not shown). At all doses examined, however, the effects of PMA on LH release were significantly lower than those of buserelin ( p < 0.01, n = 5). Aging did not affect the EDs0 of either agonist but resulted in a significant reduction of LH output by both agonists ( p < 0.01), the effects being more pronounced for buserelin than for PMA ( p < 0.05, n = 5). From these experiments we concluded that changes in the rate of LH release could be optimally studied by a 3 h incubation with 10 -7 M of agonists and all further analy~es were performed using these conditions. BUSE]~EI.IN ( . , ~ 0 = 4 4 MONItlS

PMA O ~ l : : l : 4--8 MG~HS

E 15o,

g

e-

Q) 100' 0

60 + 8% (mean + SE) of the total intracelhlar LH content measured at baseline. These effects could be mimicked through the activation of protein kinase C by an active phorbol ester (PMA, !0 -7 M) whereas incubation with buffer only or with an inactive 17~-phorbol ester (results not shown) Lad no effects oa LH secretion. As represented in Fig. 1B, an additive effect was observed by coincubating mouse pituitary cells with both buserelin and PMA at lower doses (from 10 -11 to 10 -s M) whereas at higher doses no further effects were noted over the maximal release of LH to 138 + 15 ng/106 cells/3 h with buserelin alone.

rv

50,

5

olo

-13 .

-12

-11 -10 -9 Log CONCU~I~'IION (M)

-8

-7

-6

Fig, 2. Effects of age on the dose-response curves of induced release of LH from pituitary cells. Cultured pituitary cells from intact young (4-8 months) or old (14-18 months) mice were stimulated under conditions as described in Materials and Methods with buffer alone or containing increasing doses of either buserelin or PMA. LH was measured in the extracelhlar milieu using an N I D D K anti-rat LH RIA kit. Each data point represents the mean:t: SE of five separate experiments, each performed in triplicate (where not shown, SE are smaller than the data point).

67

fi

o t--

-I_i

A-INTACT

:I_|i "l 600,

E

4--8 months

!""1: CONTROL I~1: BUSERELI~ 10-7M 1 ~ : PMA I O - I M

10-12 months

14-18 months

10-12 months

14-18 months

B-OVX

D

0

JC

!o

)

400,

tO 0 200,

&

5

n 4-8 months

Fig. 3. Effects of age and ovariectomy on the release of LH by buserelin and PMA from pituitary cells. Cultured pituitary cells obtained from tnice of all three age subgroups were stimulated for 3 h at 37°C with buffer alone (control) or containing 10-7 M of either buserelin or PMA. At the end of each stimulation period, the milieu was assayed for LH content using an NIDDK anti-rat LH RIA kit. Each bar represents the mean + SE of 8-10 separate experiments, each performed in triplicate.

Fig. 3 summarizes the secretion of LH by pituitaries of mice from all three age groups studied either intact or 14 days post-OVX. When pituitary cells from intact animals (Fig. 3A) of reproductive age were stimulated with culture medium only (control), the mean + SE LH release in the extracelhlar milieu was 36 + 16 ng/106 cells after 3 h. In contrast, when stimulations were done with 10 -? M buserelin, the rate of release increased 4-fold higher than control ones ( p < 0.01) from a baseline level of 11 + 1 n g / 1 0 e cells prior to stimulation to one of 141 + 5 ng/106 cells after 3 h ( p < 0.001). Similar experiments with 10- 7 M PMA showed that the secretion rate of LH increased to 118 + 6 ng/106 cells/3 h (p 0.05). This significant reduction in the release of LH persisted into old animals, where the following rates were observed: 6 + 1 (controls), .57+3 (buserelin) and 62+ 1 (PLIA) rig/106 cells/3 h. These effects, however, were comparable to those observed with pituitary cells from middle-aged mice ( p > 0.05). That this age-related reduction in the ability of pituitary cells to release LH cannot be accounted for by reduced cellular releasable LH (at baseline) is illustrated in Table 1. When compared to intact animals, OVX (Fig. 3B) resulted in a significant increase in basal LH TABLE 1 LH RELEASE AS A FUNCTION OF CELLULAR CONTENT (MEAN 4. SE) Cultured pituitary cells from mice of various age subgroups were stimulated with buffer (control) or 10 - 7 M of either buserelin or PMA as described in Materials and Methods. LH was then assayed by RIA of the intraceUular pool after sonication of cells as well as in the culture medium. Animal

Baseline

~ Release of total LH

status

intracelhlar LH content (ng/10 6 cells)

Control

Intact Young Middleaged Old OVX Young Middleaged Old

Buserelin

PMA

141± 6

84-1

684-11

54+ 8

1294.10, 116+ 8

74.2 7+2

674. 3 60+ 8

55-I- 3 54+ 7

3274.26 *

144-1

754- 8

734-11

271 -l-29 * 2634.16 *

10 4. 2 144.2

69 + 8 704. 9

72 + 9 70+12

* p < 0.05, intact vs. OVX

68

secretion by pituitary cells from all three age groups (p < 0.01). However, as in intact animals, the control LH release rate of 55 + 4 ng/106 cells/3 h observed in young mice was higher than those of 22 + 2 and 26 + 3 rig/106 cells/3 h measured at middle and old ages, respectively (p < 0.05). Similarly, after stimulation with buserelin or PMA, pituitary cells from OVX mice of all three age groups showed an increased ability to release LH, when compared to those from intact animals. Nevertheless, the same age-related reduced secretion rate of LH was observed, pituitary cells from OVX young mice showing a greater than 2-fold LH output after stimulation with buserelin (510 ± 40 ng/106 cells/3 h) or PMA (440 4-24 ng/106 cells/3 h) than those observed from pituitaries of middle-aged and old mice (p < 0.01).

Inositol phosphate accumulation Table 2 summarizes the incorporation of [3H]inositol into total PI, PIP, PIP2, and into total IP accumulation after stimulations with increasing doses of buserelin. As can be seen, the incorporation into total PIP as well as into PIPe remained comparable in both groups of unstimulated (controls) and stimulated pituitary cells. The observation that there is no significant reduction in phosphoinositol lipids in our experimental conditions (15 rain incubation period) suggests that the magTABLE 2 RADIOACTIVITY IN PHOSPHOINOSITOL LIPIDS Cultured cells from young mouse pituitary were labeled with [3H]myo-inositol as described in Materials and Methods. After hormonal stimulation, total phosphatidylinositol phosphates (Pl, PIP and PIP2) were determined after acid extraction and hydrolysis of cell phospholipids. Pituitary cells from old mice gave similar results. The results are expressed×10a/106 cells/15 min. Doses of Incorpo- IncorpoIncorpoIncorpobusere- rated rated rated rated lin (M) P](dpm) PIP(dpm) PIP2(dpm ) [3H]IP(dpm) 0 10 -l]

140-1-15

8.6±1.0

147±17

4.8-1-0.6 4.6±0.7 4.04-0.4

9.4±0.4 10,5 ±0.5 12.1±0.6

3.8±0.5 5.6±0.4 4.6±0.5 4.3±0.6

14.3±0.7 15.9±0.8 18.0±0.9 17.6+0.9

10 -1°

129±14

8.4±0.8 7.8±0.9

10 -9 10 -8 10 -7 10 -6

113±15 143±21 133±11 142±17

7.0+0.7 7.9±0.9 8.3±1.0 7.3±0.8

A - INTACT

X

[~:

CONTROL

ram: BUSE}~EUN

/

~, ~'~" ,F-

~

o

~

x 413. a

A

0

20-

4-8 months

N

14.-18 months

B-0VX

----. x 4, n

eo

4-8 months

14-18 months

Fig. 4. Effects of age and ovariectomy on total inositol phosphate accumulation. Cultured mouse pituitary cells were incubated as described in Materials and Methods without or with 10 -7 M buserelin for 15 mix. Total IP was then assayed as described in Table 2. Each bar represents the mean :1:SE of six separate experiments each performed in tripficate.

nitude of the hormonal effect is not fimited by a progressive reduction in the size of the hormonesensitive phosphoinositol pools. This probably reflects a rapid resynthesis of the phosphoinositol pools under hormonal stimulation. In contrast, when compared to control cells, pituitary ceils stimulated with increasing doses of buserelin showed an increased total IP accumulation which reached a maximal level of 2.2-fold over the control one at a buserelin concentration of 10 -7 M ( p