Mechanisms involved in the pituitary desensitization induced by gonadotropin-releasing hormone agonists

Mechanisms involved in the pituitary desensitization induced by gonadotropin-releasing hormone agonists Tsuguo Uemura, MD, PhD, Takashi Yanagisawa, MD, Kazuhiro Shirasu, MD, Akemi Matsuyama, MD, PhD, and Hiroshi Minaguchi, MD, PhD Yokohama, Japan OBJECTIVE: We investigated the mechanisms of desensitization induced by gonadotropin-releasing hormone agonist in the pituitary. STUDY DESIGN: Effects of gonadotropin-releasing hormone agonist on the pituitary were studied in vitro and in vivo in the rat. In the clinical study serum luteinizing hormone was measured by radioimmunoassay with a polyclonal luteinizing hormone antibody (luteinizing hormone-radioimmunoassay) and by immunoradiometric assay with monoclonal luteinizing hormone antibodies (luteinizing hormone-immunoradiometric assay) during gonadotropin-releasing hormone agonist treatment. RESULTS: In the in vitro study bead-attached pituitary cells that were desensitized with a continuous infusion of 1O- 7 mo1/L gonadotropin-releasing hormone responded to 50 mmol/L K+. In the in vivo study gonadotropin-releasing hormone binding sites and rat luteinizing hormone (3-messenger ribonucleic acid in the pituitary decreased during gonadotropin-releasing hormone agonist treatment, but serum levels of rat luteinizing hormone did not decrease. In addition, a disparity between luteinizing hormone-radioimmunoassay and luteinizing hormone-immunoradiometric assay was demonstrated during gonadotropin-releasing hormone agonist treatment. CONCLUSION: Pituitary desensitization in response to gonadotropin-releasing hormone agonist may not be wholly receptor mediated and a nonreceptor process may be involved. (AM J OBSTET GVNECOL 1992;167:283-91.)

Key words: Pituitary desensitization, gonadotropin-releasing hormone agonist, luteinizing hormone, luteinizing hormone f3-messenger ribonucleic acid, gonadotropin-releasing hormone receptor Gonadotropin-releasing hormone agonists (GnRHa) have proved useful for the treatment of endometriosis because GnRHa affect only specific organs and induce hypogonadotropic hypoestrogenism (desensitization), which renders the endometrial implant atrophic. However, the mechanisms involved in the desensitization of the pituitary are still poorly understood. To investigate these mechanisms, we administered GnRH agonist to rats and humans for more than 4 weeks. We then measured the concentrations of serum rat luteinizing hormone (LH) and pituitary rat LH, the response of serum rat LH to GnRH, and GnRH-binding sites and gene expression of rat LH 13 in the pituitary in the rat. In humans who were undergoing GnRH agonist treatment, we investigated the concentration of serum LH determined by a polyclonal antibody and by monoclonal antibodies. We also examined rat pituitary cells in vitro. From the Department of Obstetrics and Gynecology, School of Medicine, Yokohama City University. Reprint requests: Tsuguo Uemura, MD, PhD, Associate Professor, Department of Obstetrics and Gynecology, School of Medicine, Yokohama City University, 3-9, Fukuura, Kanazawa-ku, Yokohama-shi, Kanagawa Pre! 236, Japan. 610138234

Material and methods Desensitization of pituitary cells in vitro. Pituitary cells obtained from Wistar male rats were dispersed in HEPES dissociating buffer (137 mmollL sodium chloride,S mmollL potassium chloride, 0.7 mmollL disodium acid phosphate, and 25 mmollL HEPES, pH 7.3) with the use of 0.4% of collagenase type II and 10 f.Lg/ml deoxyribonuclease II (Sigma Chemical Co., St. Louis, Mo.). The final cell suspension (approximately 10 6 cells/ml) was cultured with swollen and autoclaved Cytodex beads (Pharmacia Biosystems, Piscataway, N.J.) at a concentration of 0.1 ml beads/ml of cell suspension as described by Smith and Vale. I The beadattached cells were packed into mini-columns (0.6 x 10 cm, Bio-Rad Laboratories, Richmond, Calif.) and superfused at a constant flow rate (0.13 mllmin). The cells were first superfused with GnRH and then with the test substances; effluent fractions were collected at 6-minute intervals and stored at - 20° C until assayed. Study of rats in vivo Animals and sample collection. Immature Sprague-Dawley female rats (21 days of age) were administered a sustained-release formulation of GnRH agonist (TAP144-SR; D-Leu 6 -[des-Gly'O-NH 21 GnRH-EA, Takeda

283

284

Uemura et al.

July 1992 Am J Obstet Gynecol

10

LHRH 10

,.-.. C

8

•

M

o

LHRH 10 6 M

'';:::;

u co

..~

0)

C

I

5

.....J

o

,II

120

180

240

300

360

420 mIn.

Fig. 1. Rat LH responses of bead-attached pituitary cells to 10- 6 moUL GnRH after continuous infusion of 10- 8 malll GnRH.

6

LHRH

r--.

C 0

1O- 7 M

K+ 50mM

I

~

-... 0

«I

4

......

Cl C

'-'

J: 2

..J

o /'

120

480

min.

Fig. 2. Rat LH responses of bead-attached pituitary cells to 50 mmol / L K + after continuous infusion of 10- 7 moIlL GnRH.

Chemical Industries, Ltd., Osaka, Japan) subcutaneously at a dose of 0.94 mg per rat every 4 weeks and were killed by decapitation at various time intervals after GnRH agonist i~ection. The pituitaries were harvested under sterile conditions and stored at - 70° C until subsequent analysis. Trunk blood was collected and stored at - 20° C until assayed. To monitor the pituitary response to GnRH blood samples were obtained by heart puncture under light ether anesthesia before and 30 minutes after a subcutaneous injection of 1 fLg of a native GnRH (Lutamin,

Daiichi Medical Co., Ltd., Tokyo, Japan) on days 0, 7, 14,21, and 28. GnRH TeceptoT assay. With use of a GnRH agonist (buserelin; [D-Ser (tBu 6) des-gly-NH 2'O] GnRH-ethlamide, HoechstJapan Co., Ltd., Tokyo, Japan) as a ligand, the binding capacity of the rat pituitary was examined as described previously? Assay of mt LH f3-messenger ribonucleic acid. The preparation of rat pituitary messenger ribonucleic acid (mRNA) for assay by dot blot hybridization was performed as described by Abbot et al. 3 Briefly, the pitui-

Desensitization induced by GnRH agonists

Volume 167 Number 1

285

3 LHRH ,....... c:

..g

-"... u

IBBr,cAMP

I

3mM

2

IV

Cl

c:

120

240

180

300

360

420

480

min.

Fig. 3. Rat LH responses of bead-attached pituitary cells to 8-bromo-cyclic adenosine monophosphate after continuous infusion of 10- 7 mol/L GnRH.

50 ~

00

E ...L: 00

iii ~ t:

'"

40

,,

II

,

-----+/

,

!

30

f

20

I

.:l!" •

-~

>

0

I

.r--l\ 1 '

10

,....{

o-.~

Control

-

LHRHa treated

-,_H___,.*.' o

2

3

4

8

12

Weeks of Treatment

Fig. 4. Changes in ovarian weight during GnRHa treatment (0.94 mg/4 weeks) (mean ± SE, n = 5). *P < 0.05, **P < 0.01 compared with controls.

taries were homogenized in 200 fLl of homogenization buffer and centrifuged at IS,OOOg. Proteins in the supernatant were denatured, extracted with phenol, and reextracted with chloroform. The RNA was precipitated overnight with ethanol and 3 mollL sodium acetate at - 70° C. The RNA was recovered by centrifugation at 1S,000g for 30 minutes and dissolved in sterile distilled water. A synthetic labeled oligodeoxynucleotide was produced as described by Gharib et al! RNA samples of 5 fLg each were applied to a gene hybridization mem-

brane, which was baked at 80° C for 2 hours before pre hybridization. The baked membrane was incubated at 65° C for 4 hours with 5 mlofprehybridization buffer before incubation with 50 to 100 ng of (phosphorus 32) complementary deoxyribonucleic acid for 3 hours at 60° C. After hybridization excess unbound (phosphorus 32) complementary deoxyribonucleic acid was removed by sequential washing with constant agitation in buffer. After washing, the membrane was dried, placed in contact with an X-ray film and autoradiographed for 16 to 24 hours at - 70° C. The mRNA

286

Uemura et al.

July 1992 Am J Obstet Gynecol

80

--

J

,' *' '*

60

E

'aD C

serum LH

**

3.0

t

l' 'i

40

I

20

I

\

"

I

I

\

--

E 2.0

** ** .... ~) ** ..

,, , ,

'aD C

I

...J 1.0

serum LH

~

.--.

I

0 - 0 Control \\

.---. GnRHa treated \

\ T

\6.

t-o~~:::-;:.eo

o~~~----~--~--~------

pituitary LH

3.0

pituitary LH

.--.

>-

:0 ] 2.0

-Cl.

'aD

;t

1.0

I

I ...J

\

...J

\

\\*

* *--- * ---~ --- .....

o~~~~--~~~~~------

mRNA

LH,8 mRNA

~ 100

C

0.-0,---6-0 " ....

_- ............ --

o

u

'0 o

o

50

'0

« z a::

E

~

*

50

--i'

.... *.................

*

«

o

2

3 4

5

18

Hours of Treatment

z

~

a 1-----'---'-------'-----'----'---a

3

7

14

28

Days of Treatment

Fig. 5. Serum rat LH (mean ± SE, n = 5), pituitary rat LH content (n = 5) and rat LH l3-mRNA (n = 3 to 4) from o to 18 hours after GnRH agonist treatment. Levels of rat LH l3-mRNA are expressed as a percentage of the control. *P < 0.05, **P < 0.01 compared with the control value (hour 0).

Fig. 6. Serum rat LH (mean ± SE, n = 5), pituitary rat LH content (n = 5), and rat LH l3-mRNA (n = 5) from 0 to 28 hours after GnRH agonist treatment. The levels of rat LH 13mRNA are expressed as a percentage of the controls. *P < 0.05, **p < 0.01 compared with controls.

content of each dot was determined by densitometry of autoradiograms with a densitometer (Ultrascan XL, Pharmacia LKB Nuclear, Gaithersburg, M.D.). Clinical study. A GnRH agonist (nafarelin; [D-Ala3(2-naphtyl)6] GnRH, Nippon Syntex KK, Tokyo, Japan) was administered intranasally at a dose of 400 j.Lg/day for 3 weeks to eight patients who had polycystic ovarian disease and seven infertile patients. Another GnRH agonist (TAP-144-SR) was administered subcutaneously at a dose of 3.8 mg per 4 weeks for 24 weeks to 16 eumenorrheic patients who had endometriosis. Blood samples were obtained twice a week, either every week or every 2 weeks. Serum levels of LH

were examined by radioimmunoassay (RIA) with a polyclonal antibody and by immunoradiometric assay (IRMA) wirh monoclonal antibodies. Hormone assays and statistical analyses. Serum and pituitary homogenates were assayed for rat LH by a double-antibody RIA technique with rat hormone kits supplied by the National Pituitary Agency (NIDDK, USA), and rat LH concentrations were expressed in NlH-LH-SI. "Daiichi" LH-kit (Daiichi Isotope Inc., Tokyo, Japan) was used for RIA with a polyclonal LH antibody and World Health Organization second immunoreactive plasma-human menopausal gonadotropin as the standard preparation. A "Spac-S LH-kit"

Desensitization induced by GnRH agonists

Volume 167 Number I

287



)

10

".....

E

......... \)/)

c .......

:r:

0

.J

/



E :J "-

OJ (f)

I

10

0

) -

- --

0

2

~

3

4

Weeks of Treatment

Fig. 7. Serum rat LH before and 30 minutes after native GnRH (1 fLgirat, subcutaneous) administration 0 to 4 weeks after GnRH agonist treatment in intact immature female rats (upper panel) and in GnRH agonist-treated immature female rats (lower panel) (mean ± SE, n = 5).

200

6---6 Control

A-A GnRHa treated

,

100

I

I ......

;r

o 3days 1

;~1

3

2

4

8

12

Weeks of Treatment Fig. 8. Effects of GnRH agonist treatment on pituitary GnRH receptor in female rats (mean ± SE, n = 5). **p < 0.01 compared with controls.

(Daiichi Isotope Inc.) was used for IRMA with monoclonal LH antibodies and World Health Organization first immunoreactive plasma-LH as the standard preparation. For follicle-stimulating hormone (FSH) determinations we used the FSH kit "Daiichi" for RIA with a poly clonal antibody, World Health Organization sec-

ond immunoreactive plasma-human menopausal gonadotropin and "Spac-s FSH-kit" for IRMA with monoclonal antibodies, and World Health Organization second immunoreactive plasma-human pituitary gonadotropin. Data were subjected to a one-way analysis of variance followed by the Student t test.

288

July 1992 Am J Obstet Gynecol

Uemura et al.

700

« ~

N=33 r =0.95 y =1.45+ 0.30 X P<0.05

350

0:: I ...J

LH RIA (mIU/ml) Fig. 9. Correlation between LH values determined by RIA (LH-RIA) and IRMA (LH-IRMA).

Results Desensitization of pituitary cells in vitro. Pulsatile infusion of 10- 8 mmollL GnRH induced pulsatile secretion of rat LH, but continuous infusion of 10- 8 mmoll L GnRH or 10- 7 mmoll L GnRH induced a rapid and transit increase of rat LH; thereafter the levels of rat LH decreased to the pretreatment level within 3 to 4 hours (desensitization). The desensitization was also observed when the pulse amplitude ofGnRH was more than 10- 7 mollL GnRH or when the pulse frequency was more than four times per hour. Even after this continuous infusion of 10- 8 mollL or 10- 7 mollL GnRH, the pituitary cells responded to 10- 6 mollL GnRH, 50 mmollL K +, and 3 mmollL 8-bromo-cyclic adenosine monophosphate (Figs. 1 to 3). Study in rats in vivo. During the treatment of GnRH agonist (0.94 mglrat) the ovarian weight was 17.1 ± 1.7 mg per ovary (mean ± SE) on day 0 and 18.3 ± 0.8 mg on day 3. Thereafter the weights decreased significantly to 15.2 ± 0.9 mg (control, 22.9 ± 2.3; P < 0.05) on day 7 and 9.2 ± 1.7 mg (33.2 ± 5.8; P < 0.01) on day 28 (Fig. 4). The serum concentration of rat LH increased after hour 1 of treatment (p < 0.05) and continued to rise until hour 2 (p < 0.01). Thereafter the levels decreased gradually but were higher than those measured in the intact control animals on day 3 (p < 0.01). On day 28 the concentrations had returned to the control levels. Serum rat FSH was also the same level as the control (Figs. 5 and 6). Serum rat LH response to GnRH (1 ILg) in GnRH agonist-treated rats decreased significantly (p < 0.05) on day 7 compared with the control group, and this

blunted response remained unchanged up to week 4 after GnRH agonist administration (Fig. 7). The pituitary rat LH content decreased gradually after GnRH agonist administration. Three hours after treatment the rat LH content was significantly lower than the basal level; it continued to drop slowly until day 3 and then remained unchanged up to day 28 (Figs. 5 and 6). GnRH receptors in the pituitary markedly decreased to 17.3 ± 2.6 fmollmg protein (mean ± SE) compared with the control group (123.2 ± 0.8; P < 0.01) on day 7 and remained at low levels (Fig. 8). The rat LH f3-mRNA concentrations in the GnRH agonist-treated group were not significantly different from those in the control group from 1 to 18 hours after GnRH agonist administration, but the mRNA concentrations in the GnRH agonist-treated animals from day 3 to day 28 (p < 0.05) were lower than those in the control group (p < 0.05) (Figs. 5 and 6). Clinical study GnRH agonist (nafarelin) administration to eight patients who had polycystic ovarian disease and seven infertile patients.

LH values determined by IRMA (LH-IRMA; Y) and those determined by RIA (LH-RIA; X) correlated significantly (r = 0.9472; p < 0.01) except for LH values during nafarelin treatment (Y = 1.45 + 0.3X) (Fig. 9). Nafarelin decreased LH-IRMA significantly from day 7 to day 21 (p < 0.01), but it did not decrease LH-RIA. Therefore the ratio LH-RIA to LH-IRMA (LHRIA/LH-IRMA) increased significantly from 8.4 ± 1.5 to 17.4 ± 2.3 (p < 0.05) at week 3 of nafarelin treatment (Fig. 10). The mean value of serum estradiol decreased from

Volume 167 Number I

Desensitization induced by GnRH agonists

289

- -; Polyclonal -.-; Monoclonal

** ; P
0 (%) .... c 100

+'

0

u

..... 0

'iR

'-'"

I ...J

50

**

**

O~--~------r-----------~----------r-3 7 o 14 21 Days of Treatment Fig. 10. Changes in LH-RIA, LH-IRMA, and estradiol in blood during treatment with GnRH agonist (nafarelin, 400 fJ.g/day) for 3 weeks (mean ± SE, n = 15).

44.1 ± 5.9pg/mlto 16.3 ± 2.9 pg/ml after treatment; this change was consistent with that of LH-IRMA but not with that of LH-RIA. GnRH agonist (TAP -144-SR) treatment for patients who had endometriosis. The mean LH-RIA was 20.1 ± 2.6 mIU/ml (n = 16) before treatment and did not decrease significantly at week 2 (21.3 ± 1.9), week 4 (16.4 ± 0.6), week 16 (14.3 ± 2.1), or week 20 (15.8 ± 1.8) of treatment. On the other hand, LHIRMA was 6.5 ± 1.3 mIU/ml before treatment and decreased dramatically from week 2 (3.3 ± 0.5; p < 0.05) to week 24 (0.5 ± 0.01; P < 0.001) of treatment. FSH-IRMA also decreased more than FSH-RIA, and the estradiol levels reflected the changes of LH-IRMA and FSH-IRMA more than those of LH-RIA and FSHRIA (Fig. 11).

Comment Treatment with GnRH agonist (0.94/mg/rat/28 days) did not decrease serum concentrations of rat LH, but it inhibited rat LH responses to GnRH agonist starting on day 7 and decreased ovarian weights significantly from day 7 to day 28. Moreover, we have observed that these dosages of GnRH agonist delayed vaginal opening in immature female rats (unpublished data). These findings indicate that the dosage of GnRH agonist used in this study was large enough to induce pituitary desensitization within 7 days. After treatment with GnRH agonist, the GnRH ag-

onist binding sites decreased significantly compared with those of controls. In the in vitro study desensitized gonadotropes responded to potassium, a high dose of GnRH, and a cyclic adenosine monophosphate derivative, 8-bromo-cyclic adenosine monophosphate, which could clearly induce rat LH secretion in nontreated pituitary cells. l These results suggest that the content of pituitary rat LH was not exhausted and that desensitization was mainly caused by an alteration at the receptor level. The mRNA of the rat LH ~-subunit appears to be an important rate-limiting factor in the synthesis of mature gonadotropin; therefore the changes in rat LH ~-mRNA were examined during the down-regulatory phase of GnRH agonist treatment. This study showed that reduced availability of rat LH ~-mRNA was an important factor that contributed to the depletion of rat LH content in the pituitary from GnRH-desensitized rats, although rat LH ~ gene expression was not involved in the marked reduction in pituitary content during the early stage of GnRH agonist treatment. In the clinical study different groups were given different types of agonists; therefore comparison between patient groups is not appropriate. The differences between continuous and subcutaneous administration versus intermittent intranasal administration are important, because continuous administration of GnRH agonist has been shown to result in greater inhibition of LH secretion than intermittent administration. s Moreover, the dosage of GnRH agonist (TAP-144-SR

290

Uemura et al.

July 1992

Am

J Obstet Gynecol

~Hanalog 3.8 mg/4 weeks

A I

30 E

~------

'-'

20

I

I

~

i * *

I

Y'2

I

: I

10

I--J..., I

I

* *

'I..,.

-

:

~ * * *

/

A_A ,;. 2-~-r 1:,x

* )..........1_2'

*

*

.......... ..........

r

~-~

* * *

*

* * * * 1:* * * *

--,;-a- -a- -w--

*

"'1

I

~--L- ... - -.--.7,,..r.....- - - - - -.Y- - - I

.....

"X_ ....

~

15 E

3'

10

E '-'

I

(/)

LL

5

/t'

*P<0.05 **P
~ 200 00

V.s. V.s.

aweek. aweek.

Cl.

/~£

o ~

.... 100 .....II)

* * * * * * * r"..-I * * * * * ~ *~y 2-D--n--~--D--2--D--D

W

t

a

2

'1

4 8 12 16 20 24 ·26

30

30

weeks

Weeks of Treatment Fig. 11. Changes in LH-RIA, LH-IRMA, FSH-RIA, FSH-IRMA, and estradiol in blood during treatment with GnRH agonist (TAP-144-SR 3.8 mg/4 wk, subcutaneous) for 24 weeks (mean ± SE, n = 16).

3.8 mg/4 wk, subcutaneous) was larger than that of GnRH agonist (nafarelin 400 f.\-g/day, inrranasal), because the absorption rate of intranasal administration is about 1% to 10%.6 Therefore TAP-144-SR suppressed LH-RIA and LH-IRMA more than nafarelin. In this study a disparity between LH-RIA and LHIRMA in the process of gonadotrope desensitization to GnRH was demonstrated; that is, the levels of LH-RIA did not change markedly, but LH-IRMA decreased dramatically during treatment with GnRH agonist. The changes in LH-IRMA were consistent with those of serum estradiol. The mechanisms involved in this dispa"ity after GnRH agonist treatment are uncertain, but

increases of immunoreactive molecules such as a-subunit and LH fragments might be related. These findings are consistent with the data presented in previous reports, that is, that unchanged immunoreactivity of serum LH after GnRH agonist treatment is partially caused by the cross-reactivity of free 0. in the LH-RIA7 and that common a-subunit gene expression is sustained in the presence of a pharmacologic concentration of GnRH agonist. 8 In the desensitized pituitary FSH-IRMA was also significantly suppressed from week 1 to week 24 and correlated well with estradiol levels; however, FSH-RIA suppression was significant only at weeks 2 and 3. It

Volume 167 Number 1

was reported that rat FSH j3-mRNA levels were markedly reduced when gonadotropes were desensitized by GnRH agonist and that the effects of GnRH agonist on FSH j3-nRNA levels were paralleled by its effects on LH j3-mRNA levels. 9 These data also suggested increases of immunoreactive molecules such as a-subunit and FSH fragments. These data suggested involvement of a postreceptor mechanism or non-GnRH receptor-mediated mechanisms, in addition to a receptor mechanism in desensitization, although chemical characterization should be investigated to elucidate the molecular heterogeneity observed during GnRH agonist treatment. We gratefully acknowledge receipt of the rat LH kits from the National Hormone and Pituitary Program, NIDDK. GnRH analogs TAP-144-SR, buserelin, and nafarelin were kindly provided by Takeda Chemical Industries, Ltd., HoechstJapan Co., Ltd., and Nippon Syntex KK. REFERENCES I. Smith MA, Vale WW. Superfusion of rat anterior pituitary

cells attached to cytodex beads: validation of a technique. Endocrinology 1980; 107: 1425-30.

Desensitization induced by GnRH agonists

291

2. Uemura T, Shirasu K, Matsuyama A, Namiki T, Minaguchi H. Increases in ovarian GnRH receptors by following GnRH treatment. Endocrinol Jpn 1984;31:377-86. 3. Abbot SD, Docherty K, Clayton RN. Regulation of LH mRNA levels by gonadal hormones in female rats. J Mol EndocrinoI1988;1:49-60. 4. Gharib SD, Bowers SM, Need LR, Chin WW. Regulation of rat luteinizing hormone subunit messenger ribonucleic acids by gonadal steroid hormones. J Clin Invest 1986;77:582-9. 5. Akhtar FB, Marshall GR, Nieschlag E. Testosterone supplementation attenuates the antifertility effects of an LHRH agonist in male rhesus monkeys. Int J Androl 1983;6:461-9. 6. Furr BJA, Hutchinson FG. Formulation ofluteinizing hormone releasing hormone analogues. In: Shaw RW, Marshall JC, eds. LHRH and its analogues-their use in gynaecological practice. London: Wright, 1989:49-63. 7. Meldrum DR, Tsao Z, Monroe SE, et al. Stimulation of LH fragments with reduced bioactivity following GnRH agonist administration in women. J Clin Endocrinol Metab 1984;58:755-7. 8. Lalloz MRA, Detta A, Clayton RN. Gonadotropin-releasing hormone desensitization preferentially inhibits expression of the luteinizing hormone j3-subunit gene in vivo. Endocrinology 1988;122:1689-94. 9. Rodin DA, Lalloz MRA, Clayton RN. Gonadotropin-releasing hormone regulates follicle-stimulating hormone 13subunit gene expression in the male rat. Endocrinology 1989; 125: 1282-9.