In Vitro Responses of Luteinizing Rat Granulosa Cells to Human ...

BIOLOGY

OF

REPRODUCTION

32,

In Vitro

935-945

(1985)

Responses

Cells to Human Sr.

PATRICIA

of Luteinizing

GRASSO’

and

Department Georgetown

Rat Granulosa

Thyroid-stimulating

University 3900

Hormone

THOMAS

M.

CRISP

of Anatomy

Schools Reservoir

Washington,

of Medicine Road, N. W. DC

and

Dentistry

20007

ABSTRACT The effect of human thyroid-stimulating hormone (hTSH) initial luteinization and subsequent prolactin (Prl)-medisted losa cells was studied. Granulosa cells, obtained from pregnant

on progesterone steroidogenesis mare’s serum

(P4) secretion during by cultured rat granugonadotropin (PMSG)-

treated immature female rats, were preincubated for 1, 3, 6, 12, or 24 h in control medium lacking added hormones or in medium containing 1.0 ig/ml human chorionic gonadotropin (hCG) or hTSH, and maintained subsequently for 6 days in medium containing 1.0 ig/ml bovine (bPrl). Indices of luteotropic stimulation were provided by 1) elevated P4 concentrations determined by radioimmunoassay of spent media samples; and 2) cytoplasmic lipid accumulation assessed by osmium tetroxide staining following fixation after 7 days of culture. Progesterone levels in media from cultures exposed to hCG for 24 h were twofold higher than control cultures, whereas those in media from cultures preincubated in hTSH for 24 h were fourfold higher than control levels. Cultures preincubated in 1.0 Mg/mI hCG for as little as 1 h and then maintained for 6 days in PrI secreted significandy more P4 than did control cultures also maintained with PrI for 6 days. Cultures preincubated in hTSH required a 24-h exposure before a significant increase in Prl-mediated P4 secretion was observed. Intensity of cytoplasmic osmiophilia correlated directly with P4 concentration. These results suggest that: 1) hTSH has the ability to promote P4 secretion during initial luteinization and to regulate subsequent PrI-mediated steroidogenesis by cultured rat granulosa cells; and 2) the mechanism by which hTSH stimulates PrI-mediated P4 secretion in this model system may differ from that of

hCG.

INTRODUCTION

activity,

Morphologic evidence of luteinization cultured granulosa cells in a number of (Channing, and Crisp,

of species

including monkey

the human (Channing

(Channing, rat (Crisp

1970), cow (Cirillo et al., 1969) and and Denys, 1975) has been based on

changes observable level and further

at the defined

studies

(Crisp

Channing,

Denys,

1975).

tion is

and

accompanied

by

by

rhesus

1972),

light

pig

is presumed cell

and

the transformasteroidogenic

the

available

Such

enhanced

luteinized

secretion is an that

granulosa

reservoir

accumulations

and

human been

(hCG), and have progesterone Denys,

of of

steroid

lipid

have

secretion

hFSH

chorionic correlated by

gonadotropin with increased

these

cells

(Crisp

and

1975).

Human

like Accepted November 13, 1984. Received June 26, 1984. Correspondence. Present address: Department Biology, The College of Saint Rose, 432 Western nue, Albany, New York 12203

provide

a readily

(hLH),

cells into luteal cells

increased

in

been observed in rat granulosa cells incubated in the presence of human follicle-stimulating hormone (hFSH), human luteinizing hormone

ultrastructural Crisp

results

to

with

precursors.

microscopic

1972;

Biochemically,

of follicular granulosa

1969),

which

of progestins (Channing et al., 1976). One of the hallmarks of this process increased accumulation of cytoplasmic lipid

the

thyroid-stimulating

other and

hLH,

hormone

pituitary and

glycoprotein the

homologous

(hTSH),

hormones, placental

hormone hCG, is a dimeric molecule consisting of two noncovalently linked dissimilar subunits, alpha and beta (Vaitukaitis et al., 1976;

of Ave-

935

936

GRASSO

Giudice

and

species,

are

Pierce,

the

alpha

1978).

identical

virtually

Within

subunits

and

these

are

coded

single gene that is expressed tary gland and the placenta man,

1981).

each

hormone

The

beta

and

hormonal specificity (Pierce et al., 1976). However, of

competes thyroid and

with plasma

that

unit

the

are

(Amr

by

a

of

for

to confer the

describing

in vitro

dimers

thyrotrop-

indicated

that

hCG

TSH for binding sites on human membranes (Amir et a!., 1980), sialic

acid

residues

for

this

required

et

each

studies

hCG

following resuspended

are unique thought

been

on

recent

ic activity

for

al.,

1982).

of its alpha

thyrotropic

Since

sub-

activity

hCG

and

hTSH

possess a common alpha subunit, tions prompted us to speculate

these observathat hTSH may

possess domain

and that the may also reside

in

gonadotropic regulating

the alpha

report

subunit.

the granulosa

and

biochemical of

effect

of

regulation

ated cells.

In this

gonadotropic

rat dices

activity activity

this

cell

communication,

potential

cultures,

using

evidence

luteotropic on

of

subsequent

steroidogenesis

initial

We

as inthe

describe

luteinization cultured

in

morphologic

prolactin

by

we hTSH

of luteinization

activity.

hTSH

of

and

CRISP

Plasma

given

hormones

in both the pitui(Fiddes and Good-

subunits

have

a

of

AND

its

(Prl)-medi-

rat granulosa

clot

cultures

centrifugation, in a thrombin

were the

established.

granulosa cell (Sigma Chemical

Briefly, pellet was Co., St.

Louis, MO) solution (0.006 IU thrombin/mI Medium 199 without antibiotics). Twenty zl of the thrombincell suspension were added to an equal volume of chicken plasma (Difco Laboratories, Detroit, MI) on carbon-coated glass coverslips (15 mm diameter) in multiwell tissue culture plates (Falcon Plastics, Oxnard, CA). The cell suspension and chicken plasma were mixed and a clot was allowed to form. An aliquot of the cell suspension was reserved and cell viability was determined by a modification of the Trypan blue

exclusion method (Phillips, 1973). Approximately 3.0 X iO viable cells were added to each culture well. Control culture medium contained 85% HEPESbuffered Medium 199, 15% human male serum (sterilized by Millipore filtration), and 50 IV penicillin plus 50 g streptomycin/mI (Crisp, 1977). Granulosa cells were preincubated at 36.5#{176}C in a humidified atmosphere of 95% air and 5% carbon dioxide in control culture medium or medium containing 1.0 Mg/mI hCG or 1.0 ug/mI hTSH for 1, 3, 6, 12, or 24 h. Following preincubation, the media were collected and stored at -20#{176}Cuntil assayed for progesterone content by radioimmunoassay (RIA) as previously described (Crisp, 1977). The cultures were then washed in control medium to remove alltraces of the preincubation hormone. Rinsing medium was replaced with 1.0 ml culture medium containing 1.0 Mg/mI bovine (b) PrI. Cultures were maintained in bPrl for 6 days with medium changes every 48 h. At each medium change, the spent media samples were collected and also stored at -20#{176}C until assayed for progesterone content by RIA.

Hormones MATERIALS Animal

AND

METHODS

Procedures

Immature (26-day-old) Sprague-Dawley-derived female rats (Dominion Laboratories, Dublin, VA), were used in this study. The rats were injected (s.c.) at 0700 h with 5 IV pregnant mare’s serum gonadotropin (PMSG; Calbiochem-Behring Corp., La Jolla, CA) in 0.2 ml phosphate-buffered saline (PBS) to induce

follicle cycle.

development and to initiate the first estrous Vaginal smears were taken 50 h later to confirm proestrus and the animals were killed by cervical dislocation. Tissue

Culture

bovine prolactin (NIAMDD-bPrI-6; biologic lU/mg; containing less than 0.5% growth hormone, 0.1% TSH, 0.5% FSH. and 0.5% LH by weight as determined by RIA), human chorionic gonadotropin (CR-121, biologic potency 13,450 Purified

potency

Procedures

The ovaries were removed, placed in a plastic Petri dish containing HEPES-buffered Medium 199 (Gibco Laboratories, Grand Island, NY) supplemented with 100 IU penicillin pIus 100 g streptomycin/mI, and washed and cleaned of connective tissue. Large preovulatory follicles (approximately 600-800 m in diameter) exhibiting a yellowish translucent appearance were punctured with a sterile microprobe. The granulosa cells and follicular fluid were released into the culture medium by applying gentle pressure to the follicle wall with sterile microforceps. The granulosa cell-containing culture medium was transferred to a sterile siiconized 12-mi centrifuge tube and a cell pellet was obtained by centrifugation at 600 X g for 5 mm.

30

LU/mg; 235 X LER-907), human thyroid-stimulating hormone (NLAMDD-hTSH-1-5, AFP-43 70B; biologic potency 1.5 LU/mg; containing 41 lU/mg FSH and 995 lU/mg LH as determined by RIA), human follicle-stimulating hormone (NIAMDD-hFSH-2, AFP 2844B; biologic potency 3925 lU/mg; 57 X LER-907 and containing 523 lU/mg LH activity), and human luteinizing hormone (NIAMDD-hLH-1-1, AFP-4345-B; biologic potency 12,583 lU/mg; 343.9 X LER-907 and containing 7 lU/mg hFSH, 0.029 lU/mg hTSH, less than 0.05% hGH, and 0.05% human PrI by weight as determined by RIA) were all obtained from the National Hormone and Pituitary Agency. Although hCG is not a physiologic hormone in the rat, its use in this study to substitute for rat LH is justified because: 1) it can be obtained as a more purified preparation; and 2) it has been used by other investigators in in vitro granulosa cell studies (Amsterdam et al., 1979a,b). Dose-Response

Curves

for

6TSH,

hFSH,

and

6LH

To demonstrate the dose-dependent influence of hTSH on initial luteinization and subsequent PrI-mediated steroidogenesis, granulosa cells were preincubated

GRANULOSA

for

24

0001 Mg/mi

h in graded

concentrations

of hTSH bPrl for 6 days. Mg/mi)

(1.0,

0.1,

and thereafter exposed

CELL

0.01,

or

to 0.1

Dose-response curves for hFSH and hLH were generated to assess the effects of the hFSH and hLH contaminants in the hTSH preparation used in this study and to determine the minimal concentrations of hFSH and hLH needed to induce enhanced PrI-mediated steroidogenesis in this model system. Granulosa cells were preincubated for 24 h in graded concentrations of hFSH (1.0, 0.1, 0.01, or 0.001Mg/mI) or hLH (1.0, 0.1, 0.01, or 0.001 Mg/mi) and thereafter exposed to 0.1 Mg/mI bPrl for 6 days. In addition, the possibility of synergism between the hFSH and hLH contaminants was also examined by preincubating granulosa cells for 24 h in culture medium containing both 0.04 lU hFSH and 1.26 LU hLH/ml, biopotencies equivalent to those in the hTSH preparation used in this study. cultures were also exposed to 0.1 Mg/mI bPrl for 6 days following the preincubation period. The concentration of bPrl was reduced in these experiments in order to minimize the potential effects of the contaminating glycoproteins in the bPrl prepara-

These

tion. Dose-response curves previously generated for rat (r) PrI indicated that 0.1 Mg/mI was just as effective as 1.0 Mg/mI in promoting steroidogenesis in this system (Crisp, 1977). Similar responses were observed in a number of replicate experiments using bPrl at both 1.0 and 0.1 Mg/mI (data not shown).

L

MATURATION

Statistical

937

Analysis

Granulosa cells from several animals were pooled and equal numbers of cells were exposed to various treatments within each experiment. Unless specified in figure legends, experiments were replicated at least twice with quadruplicate incubations per treatment in each experiment. Data from replicate experiments were pooled. Means and standard errors were calculated and statistical differences between treatment groups were cance

determined by analysis level was set at P