BIOLOGY OF REPRODUCTION 46, 1196-1204 (1992)

1196

Mouse Oocytes Promote Proliferation of Granulosa Cells from Preantral and AntralFollicles In Vitro1

BARBARA C. VANDERFIYDEN,3 EVELYN E. TELFER,4 and JOHN J. EPPIG2

The Jackson Laboratory, Bar Harbor, Maine 04609

ABSTRACT

Evidence is now emerging that the oocyte plays a role in the development and function of granulosa cells, This study focuses

on the role of the oocyte in the proliferation of (1) undifferentiated granulosa cells from preantral follicles and (2) more dif-

ferentiated mural granulosa cells and cumulus granulosa cells from antral follicles. Preantral follicles were isolated from 12-day-

old mice, and mural granulosa cells and oocyte-cumulus complexes were obtained from gonadotropin-primed 22-day-old mice.

Cell proliferation was quantified by autoradiographic determination of the 5H-thymidine labeling index. To determine the role

of the oocyte in granulosa cell proliferation, oocyte-cumulus cell complexes and preantral follicles were oocytectomized (OOX),

Qocytectomy being a microsurgical procedure that removes the oocyte while retaining the three-dimensional structure of the

complex or follicle. Mural granulosa cells as well as intact and OOX complexes and follicles were cultured with or without FSH

in unconditioned medium or oocyte-conditioned medium (1 oocyte/�l of medium). Preantral fofficles were cultured for 4 days,

after which 3H-thymidine was added to each group for a further 24 h. Mural granulosa cells were cultured as monolayers for an

equilibration period of 24 h and then treated for a 48-h period, with 5H-thymidine added for the last 24 h. Oocyte-cumulus cell

complexes were incubated for 4 h and then 5H-thymidine was added to each group for an additional 3-h period. FSH and/or

oocyte-conditioned medium caused an increase in the labeling index of mural granulosa cells in monolayer culture; however,

no differences were found among treatment groups. Both in oocyte-cumulus cell complexes from antral follicles and in preantral

follicles, the removal of the oocyte resulted in a decrease in the labeling index compared with intact samples (23% reduction

in OOX complexes and 71% reduction in OOX preantral follicles). Oocyte-conditioned medium reversed this effect to bring the

labeling index of OOX complexes and follicles to the level of intact samples. Oocyte-conditioned medium had no effect on intact

oocyte-cumulus cell complexes but resulted in a doubling of the labeling index in intact preantral follicles. FSH had a negative

effect on proliferation in oocyte-cumulus cell complexes, causing a reduction of 32% in intact complexes and 27% in OOX

cumulus complexes. The treatment of OOX complexes with FSH prevented the stimulatory effect of conditioned medium. No

effect of FSH was observed on preantral follicles in these short-term cultures. These results indicate that the oocytes secreted

one or more factors that promoted granulosa cell proliferation by both the relatively undifferentiated granulosa cells of preantral

ovarian follicles and the more differentiated cumulus and mural granulosa cells of antral follicles. The terminally differentiated

cumulus cells treated with FSH, however, failed to respond to the oocyte-derived proliferation factor(s); this observation indicates

that the factor can act on granulosa cells only at specific stages of their development.

INTRODUCTION

At birth the murine ovary contains only primordial fol-

licles that consist of a primary oocyte closely associated with

a single layer of squamous granulosa cells, In response to

an unidentified signal, the granulosa cells proliferate to form

a multilayered epithelium of granulosa cells around the oo-

cyte. Toward the end of the oocyte’s growth phase, the

granulosa cells differentiate into two subpopulations or-

ganized as pseudostratified epithelia [1, 21: mural granulosa

cells attached to the basement membrane enclosing the fol-

licle, and cumulus granulosa cells attached and metaboli-

cally coupled to the oocyte. Differences in secretory prod-

ucts have been found in mural and cumulus granulosa cells,

Accepted Februar 14, 1992.

Received November 22, 1991.

‘This research was supported by NIH grant HP 23839.

‘Correspondence: John J. Eppig, The Jackson Laboratory, 600 Main Street, BarHarbor ME 04609. FAX: (207) 288-5079.

3Supported by a post-doctoral fellowship from the Medical Research Council of

Canada. Present address: Department of Medicine, University of Ottawa, 451 Smvth

Road, Ottawa, Ontario, Canada K1H 8M5.

‘Supported by a post-doctoral fellowship from the Rockefeller Foundation. Pres-

ent address: Institute of Ecology and Resource Management, University of Edinburgh,

School of Agriculture, King’s Building, Edinburgh, Scotland EH9 3JG.

including a dramatic difference in their response to the

preovulatory surge of gonadotropins. Gonadotropins stim-

ulate cumulus granulosa cells to produce and secrete hy-

aluronic acid that disperses the cumulus cells, a process

called expansion or mucification [3-5]. Mural granulosa cells

do not undergo expansion, but become luteal cells. These

two cell types also differ in their distribution of receptors

[6-8] and their steroidogenic capabilities (9-111.

In experiments on intact and hypophysectomized ani-

mals, both estrogen and FSH have been found to stimulate

granulosa cell proliferation in vivo [12-14]. The mitogenic

effect of FSH is probably mediated by its stimulation of es-

tradiol production by the granulosa cells and, similarly, es-

tradiol causes induction of FSH-receptors on granulosa cells

(131, suggesting that these two hormones act synergistically

to promote granulosa cell proliferation in vivo. In numer-

ous studies, the growth, development, and function of gran-

ulosa cells have been examined by culturing the cells as

monolayers in the presence of various hormones and growth

factors; but the interpretation of these studies is compli-

cated because of the loss of the three-dimensional orga-

nization of the follicle, the disruption of the intercellular

communication normally provided by the gap junctions be-

OOCYFES PROMOTE GRANULOSA CELL PROLIFERATION 1197

tween granulosa cells, and the changes in the cytoskeletal

structure of these cells [15-17]. Nevertheless the results of

several of these studies have suggested that (1) the ovary

synthesizes epidermal growth factor (EGF) [18], somato-

medin-C/insulin-like growth factor-i (IGF-i) (19, 20], trans-

forming growth factors-a and -13 121], and fibroblast growth

factor (FGF) [22, 23], and (2) these factors, either alone or

in combination with other growth factors or gonadotro-

pins, regulate granulosa cell proliferation in many species

[24-281.

Another consideration that could potentially influence

interpretation of these studies on the regulation of granu-

losa cell proliferation in vitro is the loss of association of

these cells with the oocyte. It has been observed that cell

division occurs more frequently in the population of gran-

ulosa cells nearest the oocvte than in the more distant cells

[29-31]. An increasing body of evidence supports early ob-

servations that the oocyte may play a role in granulosa cell

development and function. The oocyte might prevent pre-

cocious luteinization of granulosa cells since removal or

death of the oocyte in situ appears to promote spontaneous

luteinization and progesterone production [32-34]. In ad-

dition, separation of the oocyte from the cumulus cells im-

pairs the ability of the cumulus cells to synthesize hyaluron-

ic acid and to undergo cumulus expansion in vitro, indicating

that mouse oocytes secrete a specific, developmentally-reg-

ulated cumulus expansion-enabling factor that allows cu-

mulus cells to undergo cumulus expansion in response to

FSH [35-37]. Using a microsurgical procedure whereby the

oocyte can be removed from an oocyte-granulosa cell com-

plex (oocytectomy), we found evidence suggesting that sol-

uble factors secreted by mouse oocvtes promote granulosa

cell proliferation and help to maintain the structural or-

ganization of the follicle [37]. This putative role of the 00-

cyte in proliferation of granulosa cells in preantral and an-

tral follicles is the subject of the present study. Three in

vitro models were used to assess the potential effect of oo-

cyte-secreted factors on granulosa cell proliferation: 1) iso-

lated preantral follicles, 2) oocvte-cumulus cell complexes

from antral follicles, and 3) mural granulosa cells from an-

tral follicles in monolayer cultures.

MATERIALS AND METHODS

Collection and Culture of Preantral Follicles

For studies using relatively undifferentiated granulosa

cells, preantral follicles were isolated from the ovaries of

12-day-old (C57BL/6J X SJL/J)F1 mice using enzymatic and

mechanical dispersion as described previously [38], with the

following modification. The ovaries were dissociated in

Waymouth MB 752/1 (WAY; Sigma Chemical Co., St. Louis,

MO) + 3 mg/mI crystallized BSA (ICN ImmunoBiologicals,

Lisle, IL) + ITS (5 p.g/ml insulin, 5 �i.g/ml iron-saturated

transferrin, and 5 �.tg/ml selenium; Collaborative Research

Inc., Bedford, MA) + 50 �tM 3-isobut I methylxanthine (IBMX;

Aldrich Chemical Co., Milwaukee, WI) + 5 mg/mI colla-

genase (Worthington Biochemical Corp., Freehold, NJ). Iso-

lated preantral follicles were washed four times in enzyme-

free medium and cultured individually for 24 h on 2% aga-

rose as described previously [37]. After overnight culture,

the oocytes were microsurgically removed from half the

follicles as described previously [35, 37]. Briefly, each fol-

licle was held with a micropipette using negative pressure.

A lancing pipette was pushed through the complex and into

the holding pipette. Upon withdrawal of the lancing pi-

pette, the negative pressure in the holding pipette aspirated

the oocyte. The resulting oocytectomized (OOX) follicle

consisted of the zona pellucida surrounded by granulosa

cells. For convenience, we refer to the non-OOX follicles

as “intact follicles” throughout this paper. It should be noted,

however, that the collagenase treatment that was used to

isolate the preantral follicles removes most of the theca cells

and components of the basal lamina. In addition, when these

follicles are cultured, the granulosa cells attach to the col-

lagen substratum and the oocyte develops within a stalk of

granulosa cells that differentiate into functional cumulus cells

within a 10-day culture period [37]; in these experiments,

however, the follicles were cultured for only 5 days.

Twenty-four-well tissue culture plates (Costar Corpora-

lion, Cambridge, MA) were coated with 250 p.1 rat tail col-

lagen prepared as described by Torrance et al. [39]. Intact

and OOX preantral follicles were cultured in 1 ml of un-

conditioned or oocvte-conditioned WAY/BSA/ITS/IBMX with

or without I p.g/ml ovine FSH-17 (NIDDK, Baltimore, MD)

in the collagen-coated wells for 4 days. Conditioned me-

dium was prepared by culturing cumulus cell-denuded fully

grown oocytes in WAY/BSA/ITS/IBMX (1 oocyte/p.l) in

uncoated wells for 24 h. Freshly prepared conditioned me-

dium was supplied to the follicle cultures every 2 days. Af-

ter the 4-day culture, 3H-thymidine (5 p.Ci/ml; Du Pont

Company, NEN Research Products, Boston, MA) was added

to each culture well for an additional 24 h. In some ex-

periments, intact or OOX follicles were grown on Costar

Transwell-COL membranes (6-mm diameter, 3.0-p.m pore

size) and were co-cultured with denuded oocvtes that were

placed either on the membrane with the follicles or under

the membrane so that contact with the granulosa cells was

prevented.

Gollection and Culture of Mural Granulosa GelLc

Mural granulosa cells were obtained from arnral follicles

of the ovaries of 22-day-old mice that had been injected 44-

48 h previously with 5 LU eCG (Diosynth, Oss, Holland).

Follicles were punctured with 25-gauge needles releasing

both oocyte-cumulus cell complexes and clumps of mural

granulosa cells. The mural granulosa cells were collected

in WAY/BSA/ITS and dispersed by being gently drawn in

and out of a Pasteur pipette, and the cell suspension was

centrifuged for 3 mm at 250 X g. The granulosa cells were

then resuspended in fresh medium and dispersed again as

1198 VANDERHYDEN ET Al..

described above. The cells were plated at subconfluent

concentrations onto fetal bovine serum (FBS)-coated Lab-

Tek chamber slides (4 chambers/slide; Nunc, mc, Naper-

ville, IL) in 400 p.1 WAY/BSA/ITS/IBMX and incubated for

24 h. After this period of equilibration, the serum-free me-

dium was changed to fresh unconditioned or oocyte-con-

diti()ned medium with or without 1 p.g/ml of FSH. The cells

were cultured for 24 h and then 3H-thymidine (5 p.Ci/ml)

was added for an additional 24 h.

Collection and Culture of Ooc-pte-Gurnulus Cell

Goinplexes

Oocyte-cumulus cell complexes were isolated from 22-

day-old mice injected 44-48 h previously with 5 IU eCG.

Complexes were isolated by puncturing the antral follicles

of the ovaries with 25-gauge needles in WAY/BSA/ITS/IBMX,

and were then washed three times in fresh medium. Com-

plexes were OOX as described previously [35]. Intact (con-

trol) and OOX complexes were cultured with or without

FSH in 200-pA drops of unconditioned or oocyte-condi-

tioned medium under oil medium or as described above.

After 4 h, 50 p.Ci/ml 3H-thymidmne was added to each group

and the complexes were incubated for a further 3 h.

Effect of Ooc-pte-Conditioned Media on 3T3 Cells and

Sertoli Cells

Mouse 313 fibroblasts and Sertoli cells were grown as

monolayers in serum-free medium in the presence or ab-

sence of oocyte-conditioned media and/or FSH. NIH mouse

3T3 cells were plated onto FBS-coated Lab-Tek tissue cul-

ture chambers for 8 h, then treated with conditioned me-

dium and/or FSH for 24 h. 3H-Thymidine was added for 3

h at the end of the treatment period. Sertoli cells were ob-

tained by dissociating testes of 6-day-old males in 1 mg/mI

of collagenase to loosen the seminiferous tubules. The tu-

bules were further dissociated with trypsin EDTA to obtain

a single cell suspension. After overnight culture in FBS-coated

Lab-Tek tissue culture chambers in serum-free medium, the

cells were washed thoroughly to remove germ cells and

then were treated with either control or oocyte-condi-

tioned media and/or FSH for 24 h. 3H-Thvmidine was added

for the final 3 h of the treatment.

Quan4flcation of Cell Pro4feration by Autoradiograpby

At the end of the labeling period, preantral follicles or

the oocyte-cumulus cell complexes were rinsed four times

with fresh WAY/BSA/ITS/IBMX and each treatment group

was transferred in a minimum volume to a well in a Ter-

asaki plate (Nunc). A 10-p.l drop of collagen solution was

added immediately and allowed to gel. The gel drops were

fixed in Bouin’s for 2-4 h, embedded in paraffin, and then

sectioned at 7-p.m thickness. Mural granulosa cell mono-

layers, 3T3 cells, and Sertoli cells were rinsed with PBS,

fixed for 20 mm in cold methanol, rinsed with PBS, and

air-dried. Autoradiographs were prepared using Kodak

(Rochester, NY) NTB2 photographic emulsion and the

emulsion-coated slides were exposed at 4#{176}Cfor 4 days. Af-

ter developing, the labeled and unlabeled cells were counted

using an Olympus bright-field microscope (40x objective).

In the granulosa cell monolayer experiments, at least 9 rep-

licates were analyzed with a minimum of 1 000 cells counted

per replicate. In experiments using complexes and prean-

tral follicles, the cells were counted in the fields showing

the largest diameter. A minimum of 1 500 cells were counted

per treatment. Each field counted contained 1-4 com-

plexes or follicles.

Statistical Analysis

Experiments were performed a minimum of three times.

The thymidmne labeling index (the percentage of cells la-

beled) was determined for each field examined. Signifi-

cance of differences between each treatment group was de-

termined by Chi-square analysis s�vi,th Yates’ correction;

differences were considered significant at p < 0.01. To

present the variability between experiments, data are shown

as the mean ± SEM.

RESULTS

In Vitro Grout/i of Intact (Control) and OOX Preantral

Follicles

Intact preantral follicles grew significantly during a 5-day

culture period but OOX complexes were smaller (Fig. 1).

Addition of oocytes (1 oocyte/p.l) to the culture medium

resulted in a marked enlargement of both control and OOX

follicles (Fig. 1). In some experiments, the oocytes were

cultured below the Costar Transwell-COL membranes so

that contact with the cultured follicles was prevented. Sim-

ilar effects of oocvtes on granulosa cell proliferation were

observed whether the denuded oocytes were cultured on

the membrane with the follicles or below the membrane,

although the effect appeared to be somewhat less when the

oocvtes were below the membrane (not shown). The ef-

fects of oocyte-conditioned medium on granulosa cell pro-

liferation were quantified by autoradiographic analysis (Fig.

2). The qualitative observations on the effects of oocyte-

conditioned medium on granulosa cell proliferation were

confirmed by the finding that 14.0 ± 1.9% of cells of intact

complexes had incorporated 3H-thymidine during the la-

beling period while removal of the oocyte resulted in a

71% decrease (p < 0.01) in the labeling index compared

with intact samples (Fig. 3). Oocyte-conditioned medium

reversed the effect of oocytectomy and restored the label-

ing index of OOX follicles to intact levels (11.0 ± 1.3%).

In addition, culture of intact complexes in oocyte-condi-

tioned medium resulted in almost a doubling of the label-

ing index to 26.0 ± 2.3%. The addition of FSH to the cul-ture medium had no effect on the labeling index of any of

the groups.

C.. � :� .�-‘4

‘ :.�..,.

#{149}‘-.�! .)b’��

� #{149}:‘_�#{149}i

U -#{149}�.;�

#{149} .#{149} #{149}4�q4- .4 “�#{149} . . . #{149}.�.#{149}‘#{149}1� #{149}#{149} #{149}. #{149}#{149}.#{149}?�I . #{149}f, #{149}, #{149},#{149}#{149},#{149}.�C,

#{149}. . #{149}- #{149}.�,... .. �- #{149}#{149}.,(#{149} � .)� I’ - #{149}#{149}#{149}�)#{149}

I #{149}. .;. ‘I #{149}�, #{149} ,�,, � �,#{149}#{149}#{149} - -#{149} - ‘ #{149}‘ :. U;�

p #{149}c7� .L” �‘,., ,. . ..#{149}.‘. ‘.

- . 14 -#{149} #{149}#{149}#{149}#{149}#{149}C

� r .#{149} . -#{149}. I

#{149}:-,., �.-�.

#{149}�,:.

D

‘15

�

0__ �0- �- U)

5)

‘00

�00

0CD0

‘0 C5)0

200�0

� 0.o�2

COWCD 0.

CD .9 -

�00(0U,

OCDO

c-C

i2 �

�

#{149}5�o�U)

-CD15

c E�__.C0 C

- (U�0

CS U) U) .0>- C CDto 5)

‘0 ‘0 0

“‘U)0.-U)OC

2 .2

00 C-�C � CD --� a,g� �

‘0

� E- 5)

c-j��CD0 (0 0

9.� E,�= C CDCD��5)

C CDU)CD CD ,�,�

I� CD

o O#{149}� 0

C �

�O�E‘-‘0 �w0) CD 0

U) �2.9 U) 5, -

-j� E.�CD a

(0 E .c 0a� o� �)

0 .... 0aCD 0 C

5)

o CD� �� 0 �#{149}CCD E CD

-CD

U) CD0) CD U) .�o ‘� 0 U)

2- �cODE0 � 0

U Z_C CD.C

�_C 00

CD .C’0u. .9 a CD

- ‘0

�- �A ‘.� ‘-� � �

- -�e�

p.‘U

U

C

#{149}�

:.. ‘�

- .. .. - ‘

:. �- ‘ #{149}

‘C.,

..j#{149}- -�.,,:

I. #{149}

5: -

::

I

OOCYTES PROMOTE GRANULOSA CELL PROLIFERATION 1199

A B

-.-�-_--�...

‘I’I-

FIG. 2. Autoradiographs of preantral follicles grown on rat tail collagen for 5 days and labeled with 3H-thymidine.IA) Intact follicle. IS) Intact follicle cultured in oocyte-conditioned medium. (C) OOX follicle. ID) OOX follicle cultured

in oocyte-conditioned medium. All of the micrographs are the same magnification; the bar indicates 100 �m.

D

#{149}Intact

000x

Control

Oocyte-Condltloned k�._...

1200 VANDERHYDEN ET AL.

U,CD

‘5,C

0)C

2

FIG. 3. Incorporation of ‘H-thymidine by complexes from preantral fol-licles during a 24-h incubation. Each bar represents the mean ± SEM ofthe labeling indexes (percentage labeled cells) for intact and OOX folliclesin four treatment groups. There were at least 20 follicles in each group.

Oocytectomy resulted in a decrease in the labeling index of preantral fol-

licles in all culture conditions (p < 0.01). Oocyte-conditioned medium stim-

ulated granulosa cell proliferation in both control and OOX complexes ineither the absence or presence of FSH (p < 0.01).

�.. -y

In Vitro Pro4feration of Mural Granulosa Cells from

Antral Follicles

To investigate a possible role for the oocyte in the reg-

ulation of proliferation of more differentiated granulosa cells,

mural granulosa cells were isolated from antral follicles and

cultured as a monolayer for 48 h in unconditioned medium

or medium conditioned by denuded oocytes for 24 h. Dur-

ing the labeling period (the latter 24 h of culture), 11.6 ±

0.6% of unstimulated cells had incorporated 3H-thymidine

(Fig. 4). The presence of oocyte-conditioned medium re-

sulted in a 41% increase (p < 0.01) in the thymidmne la-

beling index of mural granulosa cells. FSH also stimulated

an increase in the thymidine labeling index of those cells

to 14.9 ± 1.4% (p < 0.01), but had no effect on the en-

hancement of labeling index induced by oocvte-condi-

tioned medium.

In Vitro Growth of Intact and OOX Complexes from

Antral Follicles

To ascertain the role of the oocvte in the proliferation

of cumulus granulosa cells, the labeling indexes of intact

Control ESH Control FSH

35-

OOCYI’ES PROMOTE GRANULOSA CELL PROLIFERATION 1201

18

16

14

12-a

10at

C)

C) 6�1

4

2

I)

Oocyte-Conditioned Medium

FIG. 4. Incorporation of ‘H-thymidine by mural granulosa cells from

antral follicles during 24-h incubation in monolayer culture. Each bar rep-

resents the mean ± SEM of labeling indexes in nine independent experi-

ments. An increase in the labeling index was observed in all treatment groups

when compared to the control group (p < 0.01), but no differences were

found among treatment groups.

and OOX oocvte-cumulus cell complexes during a 3-h la-

beling period were determined. Oocytectomy resulted in

a decrease in the index from 32.4 ± 1.3% in intact com-

plexes to 24.8 ± 1.6% (p < 0.01; Fig. 5). The labeling index

was restored to that of intact complexes when the OOX

complexes were cultured in oocyte-conditioned medium.

Oocvte-conditioned medium had no effect on intact oocvte-

cumulus cell complexes.

Fewer cumulus cells were labeled in the complexes cul-

tured in the presence of FSH than in unstimulated com-

plexes (22.1 ± 1.2% vs. 32.4 ± 1.3%;p < 0.01). Ooctec-

tomy further reduced the labeling index of FSH-treated

complexes to 18.0% ± 1.3 (p < 0.01). FSH also caused a

similar decrease of the index in both intact (53% decrease)

and OOX (41% decrease) complexes cultured in oocyte-

conditioned medium.

Neither oocyte-conditioned media nor FSH had an effect

on the proliferation of mouse 3T3 cells or Sertoli cell

monolayers (data not shown).

DISCUSSION

That the oocvte may play a role in proliferation of the

granulosa cell layers was suggested by earlier studies wherein

3H-thymidine was infused in vivo and labeling of the gran-

ulosa cells closest to the oocyte was greater than labeling

of those closest to the basement membrane [30, 31]. Two

results presented here clearly demonstrate that the oocyte

promotes the proliferation of granulosa cells from both

preantral and antral follicles. First, oocvtectomy resulted in

a decrease in cell proliferation and second, medium con-

ditioned by oocytes restored proliferation at least to the

level of the intact controls. In addition, medium condi-

tioned by oocytes promoted granulosa cell proliferation even

in intact preantral follicles and in monolayers of mural

granulosa cells from antral follicles. Although the oocyte is

normally coupled with its companion granulosa cells via

gap junctions 140-42], the action of the oocyte on granulosa

cell proliferation does not require contact of the oocyte with

granulosa cells. The oocyte, therefore, stimulates granulosa

cell proliferation by the production and secretion of one

or more soluble growth factors.

The oocyte-derived growth factor is not yet character-

ized. Although this factor had no detectable effect on the

proliferation of 3T3 cells or Sertoli cells, it is not known

whether it is a commonly recognized “growth factor”,

whether it is produced by other cell types, or whether its

activity is specific to granulosa cells. Nevertheless, it is

probably not epidermal growth factor or transforming growth

factor-a, since these are potent stimulators of cumulus ex-

pansion [43, 44] and oocyte-conditioned medium does not

stimulate cumulus expansion in the absence of FSH [35,

44]. The oocvte-derived growth factor is probably distinct

from the oocvte-derived cumulus expansion enabling fac-

tor. The enabling factor is secreted only by germinal vesicle

breakdown (GVB )-competent oocytes, both meiosis-ar-

rested and mature, while the proliferation factor is secreted

by both GVB-competent and incompetent oocytes [37]. IBMX

was used in the present study on the proliferation factor to

maintain oocvtes in meiotic arrest throughout the culture

periods. However, it is not known whether maturing or ma-

ture oocytes secrete the proliferation factor.

The growth of preantral follicles is thought to occur in-

dependently of gonadotropins since these follicles have been

reported to develop even in hvpophysectomized animals

[45, 46], in a gonadotropin-deficient mutant [47], or when

circulating gonadotropins have been immunologically neu-

tralized [48]. Local autocrine and paracrine growth factors,

US

CD

C

at

C

2C)-J

Oocyte-Conditioned Medium

FIG. 5. Incorporation of 3H-thymidine by intact and OCX oocyte-cu.mutus cell complexes from antral follicles in four treatment groups. Barsrepresent the mean ± SEM of the labeling indexes with a minimum of 20

complexes per treatment group. Oocytectomy resulted in a decrease in thelabeling index in unstimulated and FSH-stimulated cultures (p < 0.01). Thelabeling index of OOX complexes was restored to intact levels when they

were cultured in oocyte-conditioned medium.

1202 VANDERHYDEN ET AL.

therefore, probably play an important role in the regulation

of preantral follicle development, although gonadotropmns

may enhance the actions of these factors and are necessary

for further antral follicle development. We did not detect

an effect of FSH on the growth of preantral follicles during

the period studied here, Moreover, the reduction of gran-

ulosa cell proliferation resulting from oocytectomy was not

reversed by FSH. Nevertheless, FSH does enhance mouse

follicular growth and development in vitro, although its ef-

fects are not obvious until later in the culture period [49].

Others have reported that FSH stimulates granulosa cell

proliferation in preantral follicles isolated from cycling

hamsters [50, 51]. Whether these divergent results are due

to differences in species, stages of sexual maturity, or meth-

odology remains to be resolved.

Once the follicle reaches the antral stage of develop-

ment, two subpopulations of granulosa cells can be iden-

tified: mural and cumulus granulosa cells. There is much

evidence demonstrating that these subpopulations have ac-

quired unique characteristics as a result of their differen-

tiation. Heterogeneity of morphology and function occurs

not only between mural and cumulus granulosa cells, but

within the mural granulosa cell subpopulation as well [6,

10, 11, 52-56]. Mural granulosa cells are not terminally dif-

ferentiated, but will become so when they undergo lutein-

ization after the preovulatory W surge. Mouse cumulus

granulosa cells also become terminally differentiated dur-

ing cumulus expansion after LH treatment in vivo or after

stimulation with FSH, but not with LH, in vitro [57]. For

these reasons, it is perhaps not surprising that the prolif-

eration of mural and cumulus granulosa cells is affected

differently by FSH in vitro, with proliferation being stimu-

lated in mural granulosa cells and suppressed in cumulus

granulosa cells. In support of this observation, cumulus cells

isolated from the oviducts of rats were found to be non-

proliferating as demonstrated by the virtual absence of cells

in 5, G2-, and M-phases [581. Furthermore, in a hypophy-

sectomized rat model, FSH and estradiol initiallystimulated

granulosa cell proliferation, but continued treatment re-

sulted in a decreased labeling index [59]. As reported from

that study, administration of LH to rats initially treated with

FSH shut down the proliferation of mural granulosa cells

in vivo [591. Thus the ability of FSH or LH to stimulate or

suppress granulosa cell proliferation was dependent upon

their stage of development. In our study, FSH stimulated

the proliferation of mural granulosa cells isolated from the

large antral follicles of eCG-primed mice and cultured in

monolayers. It is possible that the lack of a normal three-

dimensional organization, changes in cell structure in-

curred by flattening in the culture dish, and the lack of con-

tact with appropriate components of basal lamina may have

influenced the response of the granulosa cells to FSH (i.e.,

proliferation rather than terminal differentiation). Compo-

nents of basal lamina affect granulosa cell shape [17, 60]

and augment W-induced cell differentiation [15, 611.

Mural granulosa cells cultured as monolayers in oocyte-

conditioned medium showed an increase in labeling index

compared to those cultured in unconditioned medium. Thecells in situ are relatively distant from the oocyte; they

maintain a physical association with it only through the gap

junctions that functionally link granulosa cells to one an-

other and to the oocyte [40-421. However, the observation

that an oocyte-secreted factor can stimulate the prolifera-

tion of mural granulosa cells suggests that the oocyte may

be secreting a granulosa cell growth factor that could act

on the mural granulosa cells, perhaps via the follicular fluid,

in a paracrine fashion.

Oocytectomy of oocyte-cumulus cell complexes does not

prevent FSH-stimulated elevation of intracellular cAMP lev-

els [35], nor does it prevent the FSH-induced cessation of

cumulus cell proliferation (this study). FSH not only shuts

down the proliferation of cumulus cells in intact com-

plexes, it also turns off the proliferation of cumulus cells

in OOX complexes cultured in oocyte-conditioned me-

dium. Thus the terminal differentiation of cumulus cells in-

duced by FSH in vitro alters the responsiveness of those

cells to the oocyte-derived growth factor(s); this observa-

tion indicates that the factor can act on granulosa cells only

at specific stages of their development,

That the mechanisms regulating follicular growth and

differentiation are very complex is becoming increasingly

clear as the potential participation of more and more fac-

tors in these processes is revealed. In addition to FSH, ste-

roids [12], theca cell-produced factors [62-64], and numer-

ous other growth factors [24, 25, 62, 63, 65-67] influence

the proliferation of granulosa cells. To further complicate

our understanding of the growth regulatory mechanisms,

gonadotropins and growth factors modulate the activity of

one another in what appears to be a delicately balanced

multifactorial system [26, 65, 68-74], Appreciation of the

oocyte as a source of additional growth factors for granu-

losa cells suggests that studies using cultures of granulosa

cells separated from oocytes may not precisely reflect the

relative roles of the myriad factors regulating proliferation

and differentiation in vivo. Conditions for maintaining long-

term cultures of complexes and follicles have been de-

scribed for several species [39, 51, 75-77]. These culture

systems support a three-dimensional organization of the

granulosa cells such that their interactions with oocytes, ex-

tracellular matrix, and one another can be maintained. The

technique of oocytectomy and the ability to grow OOX

granulosa cell complexes in vitro provide the opportunity

for further investigations of the role of the oocyte in gran-

ulosa cell proliferation as well as the interactions between

the oocyte and the various other factors such as hormones,

growth factors, and extracellular matrix that have been im-

plicated in the regulation of granulosa cell proliferation.

ACKNOWLEDGMENTS

The authors thank Philip J. Caron for his technical assistance. The FSH used in

this study was generously provided by the NIDDK through the National Hormone

and Pituitary Program (University of Maryland School of Medicine, Baltimore, MD).

OOCY�ES PROMOTE GRANULOSA CELL PROLIFERATION 1203

REFERENCES

I. Anderson E. Wilkinson RF, Lee G. Meller S. A correlative microscopical analysis

of differentiating ovarian follicles of mammals. J Morphol 1978; 156:339-366.

2. Lipner H, Cross NL. Morphology of the membrana granulosa of the ovarian fol-

licle. Endocrinology 1968; 82:638-641.

3. Dekel N, Kraicer PF. Induction in vitro of mucifIcation of rat cumulus oophorus

by gonadotropins and adenosine 3’,S’-monophosphate. Endocrinology 1978;

102:1797-1802.

4. Eppig jI. FSH stimulates hvaluronic acid synthesis by oo te-cumulus cell com-

plexes from mouse preovulatorv follicles. Nature 1979; 281:483-484.

5. Salustri A, Yanagishita M, Hascall VC. S nthesis and accumulation of hyaluronic

acid and proteoglycans in the mouse cumulus cell-oocyte complex during fol-

licle-stimulating hormone-induced mucification. J Biol Chem 1989; 264:13840-

13847.

6. Amsterdam A, Koch Y, Lieberman ME, Lindner HR Distribution of binding sites

for human chorionic gonadotropin in the preovulatory follicle of the rat.J Cell

Biol 1975; 67:894-900.

7. Bortolussi M, Marini G, Reolon ML. A histochemical study of the binding of 2tI�

labelled HCG to the rat ovars- throughout the estrus cycle. Cell Tissue Res 1979;

197:213-226.

8. Lawrence IS, Dekel N, Beers WI-I. Binding of human chorionic gonadotropin by

rat cumuli oophori and granulosa cells: a comparative study. Endocrinology 1980;

106:1114-118.

9. Hillensjo T, Magnusson C, Svensson U, Thelander H. Effect of Li-I and FSH on

steroid synthesis by cultured rat cumulus cells. In: Schwartz NB. Hunzicker-Dunn

M (eds.), Dynamics of ovarian Function. New York: Raven Press; 1981: 105-110.

10. Zoller LC, Weisz J. Identification of cvtochrome P-450, and its distribution in the

membrana granulosa of the preovulatorv follicle, using quantitative cytochem-

istrv. Endocrinology 1979; 103:310-313.

11. Zoller LC, Weisz J. A quantitative cytochemical study of glucose-6-phosphate de-

hydrogenase and 5-3(3-hydroxysteroid dehydrogenase activity in the membrana

granulosa of the ovulable type of follicle of the rat. Histochemistrv 1979; 62:125-

135.

12. Goldenberg RL, Vaitukaitis JL, Ross GT. Estrogen and follicle-stimulating hor-

mone interactions on follicle growth in rats. Endocrinology 1972: 90:1492-1498.

13. Louvet J, Vaitukaitis JL Induction of follicle-stimulating hormone (FSH) receptors

in rat ovaries by estrogen priming. Endocrinology 1976; 99:758-764.

14. Richards iS. Hormonal control of ovarian follicular development: a 1978 per-

spective. Recent Prog I-form Rca 1979; 35:343-373.

15. Amsterdam A, Rotmensch S. Furman A, Venter EA, Vlodavasky I. S�ergistic effect

of human chorionic gonadotropin and extracellular matrix on in vitro differ-

entiation of human granulosa cells: progesterone production and gap junction

formation. Endocrinology 1989; 124:1956-1964.

16. Ben-Rafael Z, Benadiva CA, Matroianni L Jr. Garcia CJ, Minda JM, lozzo RV, Flick-

inger GL. Collagen matrix influences the morphological features and steroid se-

cretion of human granulosa cells. Am J Obstet Gynecol 1988; 159:1570-1574.

17. CarnegiejA, Byard R, Dardick I, Tsang BK. Culture of granulosa cells in collagen

gels: the influence of cell shape on steroidogenesis. Biol Reprod 1988; 38:881-

890.

18. RaIl LB, Scott J, Bell GI, Crawford RJ, Niall HD, Cogclan JP. Mouse prepro-epi-

dermal growth factor synthesis b�’ the kidney and other tissues. Nature 1985;

313:228-231.

19. Davoren JB, Hsueh AJW. Growth hormone increases ovarian levels of immu-

noreactive somatomedin C/insulin-like growth factor I in vivo. Endocrinology

1986: 118:888-890

20. Hammond JM, Baranao JLS, Skaleris D, Knight AB, Romanus JA, Rechler M.M.

Production of insulin-like growth factors by ovarian granulosa cells. Endocri-

nology 1985; 117:2553-2555.

21. Skinner MK. Transforming growth factor production and action in the ovarian

follicle: theca cell-granulosa cell interactions. In: Hirshfield AN (ed). Growth

Factors and the Ovary. New York: Plenum Press; 1989: 141-150.

22. Koos RD. Olson CE. Expression of basic fibroblast growth factor in the rat ovary:

detection of mRNA using reverse transcription-polymerase chain reaction am-

plification. Mol Endocrinol 1989; 3:2041-2048.

23. Neufeld G, Ferrara N, Schweigerer L, Mitchell R, Gospodarowicz D. Bovine gi-an-

ulosa cells produce basic fIbroblast growth factor. Endocrinology 1987; 121:597-

603.

24. Gospodarowicz D, Ill CR, Birdwell CR. Effects of fibroblast and epidermal growth

factors on ovarian cell proliferation in vitro. 1. Characterization of the response

of granulosa cells to FGF and EGF. Endocrinology 1977; 100:1108-1120.

25. Gospodarowicz D, Bialecki H. Fibroblast and epidermal growth factors are mi-

togenic agents for cultured granulosa cells of rodent, porcine and human origin.

Endocrinology 1979; 104:757-764.

26. Hammond JM, Mondschein J5, Canning SF. Insulin-like growth factors (IGFs) as

autocrine/paracrine regulators in the porcine ovarian follicle. In: Hirshfield AN

(ed), Growth Factors and the Ovary. New York: Plenum Press; 1989: 107-120.

27. Schomberg DW. Growth factor-gonadotropin interactions in ovarian cells. In:Hirshfleld AN (ed.), Growth Factors and the Ovary. New York: Plenum Press;

1989: 131-139.28. MayJV. Multifactorial regulation of granulosa cell (GC) proliferation: interactions

among polypeptide growth factors. In: Hirshfield AN (ed), Growth Factors and

the Ovary. New York: Plenum Press; 1989: 227-232.

29 Bullough WS. The method of growth of the follicle and corpus luteum in themouse ovary. J Endocrinol 1942; 3:150-156.

30. Hirshfield AN. Patterns of [3H)thymidine incorporation differ in immature ratsand mature, cycling rats. Biol Reprod 1986; 34:229-235.

31. Takaoka H, Satoh H, Makinoda 5, Ichinoe K. Granulosa-cell growth factor in oo-

cyte and its transport systems. Acts Obstet Gvnaecol Jpn 1985; 37:92-98.

32. El-Fouly MA, Cook B, Nekola M, Nalbandov AV. Role of the ovum in follicular

luteinization. Endocrinology 1970; 87:288-293.

33. Hubbard GM, Erickson GF. Luteinizing hormone-independent luteinization and

ovulation in the hypophysectomized rat: a possible role for the oocyte. Biol Re-

prod 1988; 39:183-194.34. Nekola MV, Nalbandov AV. Morphological changes of rat follicular cells as influ-

enced by oocytes. Biol Reprod 1971; 4:154-160.35. Buccione It, Vanderhyden BC, Caron PJ, Eppig B. FSH-induced expansion of the

mouse cumulus oophorus in vitro is dependent upon a specific factor(s) se-

creted by the oocyte. Dcv Biol 1990; 138:16-25.

36. Salustri A, Yanagishita M, Hascall VC. Mouse oocvtes regulate hyaluronic acidseothesis and mucification by FSH-stimulated cumulus cells. Dcv Biol 1990; 138:26-

32.

37. Vanderhyden BC, Caron PJ, Buccione R, Eppig B. Developmental pattern of the

secretion of cumulus-expansion enabling factor by mouse oocytes and the roleof oocvtes in promoting granulosa cell differentiation. Dcv Biol 1990; 140:307-

317.

38. Eppig B. Downs SM. The effect of hypoxanthine on mouse oocyte growth anddevelopment in vitro: maintenance of meiotic arrest and gonadotropin-inducedoocvte maturation. Dcv Biol 1987: 119:313-321.

39. Torrance C, Telfer E, Gosden RG. Quantitative study of the development of iso-

lated mouse pre.antral follicles in collagen gel culture. J Reprod Fertil 1989;

87:367-374.

40. Aibertini DF, Anderson E. The appearance and structure of the intercellular con-

nections during the ontogeny of the rabbit ovarian follicle with special reference

to gap junctions. J Cell Biol 1974; 63:234-250.

41. Anderson E, Albertini DF. Gap junctions between the oocyte and companionfollicle cells in the mammalian ovary. J Cell Biol 1976; 71:680-686.

42. Gilula NB, Epstein ML, Beers WH. Cell-to-cell communication and ovulation. Astudy of the cumulus-oocvte complex. J Cell Biol 1978; 78:58-75.

43. Downs SM. Specificity of epidermal growth factor action on maturation of the

murine oocvte and cumulus oophorus in vitro. Biol Reprod 1989; 41:371-379.

44. Salustri A, Ulisse 5, Yanagishita M, Hascall VC. Hyaluronic acid synthesis by mural

granulosa cells and cumulus cells in vitro is selectively stimulated by a factor

produced by oocytes and by transforming growth factor �. J Biol Chem 1990;

265:19517-19523.

45. Moore PJ, Greenwald GS. Effect of hypophysectomv and gonadotropin treatment

on follicular development and ovulation in the hamster. AmJ Anat 1974; 139:37-

48.

46. Smith PE. Hypophysectomy and a replacement therapy in the rat. Am J Anat 1930;

45:205-273.

47. Halpin DMG, Jones A, Fink G, Charlton HM. Postnatal ovarian follicle develop-ment in hvpogonadal (hpg) and normal mice and associated changes In the hy-pothalamic-pituitary ovarian axis. J Reprod Fertil 1986; 77:287-296.

48. Eshkol A, Lunenfeld B, Peters H. Ovarian development in infant mice. Depen-

dence on gonadotropic hormones. In: Butt WIt. Crooke AC, Rvle M (eds.), Go-

nadotrophins and Ovarian Development; Edinburgh: E & S Livingstone; 1970:249-258.

49. Eppig B. Maintenance of meiotic arrest and the induction of oocyte maturation

in mouse oocvte-granulosa cell complexes developed in vitro from preantral

follicles. Biol Reprod 1991; 45:824-830.50. Roy SK, Greenwald G5. Effects of FSH and LH on incorporation of (3H)thymidine

into follicular DNA. J Reprod Fertil 1986; 78:201-209.

51. Roy SK, Greenwald GS. Hormonal requirements for the growth and differentia-tion of hamster preantral follicles in long-term culture. J Reprod Fertil 1989;

87:103- 114.52. Erickson GF, Hofeditz C, Unger M, Allen WIt, Dulbecco R A monoclonal antibody

to a mammary cell line recognizes two distinct subtypes of ovarian granulosa

cells. Endocrinology 1985; 117:1490-1499.

1204 VANDERHYDEN ET AL.

53. Hartshorne GM. Subpopulations of granulosa cells within the human ovarian

follicle. J Reprod Fertil 1990: 89:773-782.

54. Rao IM, Mills TM, Anderson E, Mahesh \‘B. Heterogeneity in granulosa cells of

developing rat follicles. Mat Rec 1991; 229:177-185.

55. Ueno 5, Takahashi M, Manganaro IF, Ragin RC, Donahoe Pl( Cellular localization

of Mullet-ian inhibiting substance in the developing rat ovary. Endocrinology 1989;

124:1000-1006.

56. Zeleznik AJ, Midgley AR Jr., Reichert J LE. Granulosa cell maturation in the rat:

increased binding of human chorionic gonadotropin following treatment with

follicle-stimulating hormone in vivo. Endocrinology 1974; 85:818-825.

57. Eppig B. Regulation of cumulus oophorus expansion by gonadotropins in vivo

and in vitro. Biol Reprod 1980; 23:545-552.

58. Schuetz AW, Whittingham DG, Legg RE. Alterations in the cell cycle characteristics

of granulosa cells during the periovulatory period: evidence of ovarian and ovi-

ductal influences. J Exp Zool 1989: 249:105-110.

59. Rao MC, Midgley AR Jr., Richards J5. Hormonal regulation of ovarian cellularproliferation. Cell 1978; 14:71-78.

60. Ben-Ze’ev A, Amsterdam A. Regulation of cytoskeletal proteins involved in cell

contact formation during differentiation of granulosa cells on extracellular ma-trix. Proc Nat Acad Sci USA 1986; 83:2894-2898.

61. Furman A, Rotmensch S, Kohen F, Mashiach S, Amsterdam A. Regulation of rat

granulosa cell differentiation by extracellular matrix produced by bovine corneal

endothelial cells. Endocrinology 1986; 118:1878-1885.

62. Bendell B. Lobb DK, Chums A, Gysler M, Dorrington JH. Bovine thecal cellssecrete factor(s) that promote granulosa cell proliferation. Biol Reprod 1988;

38:790-797.

63. Makris A, Klagsbrun MA, Yasumizu I, Ryan I�. An endogenous ovarian growthfactor which stimulates BALB/3T3 and granulosa cell proliferation. Biol Reprod

1983; 29:1135-1141.

64. Skinner MK, Lobb D, Dorrington Jil. Ovarian thecal/interstitial cells produce an

epidermal growth factor-like substance. Endocrinology 1987; 121:1892-1899.

65. May JV, Frost JP, Schomberg DW. Differential effects of epidermal growth factor,

somatomedin-C/insulin-like growth factor-I, and transforming growth factor-u

on porcine granulosa cell deoxyribonucleic acid synthesis and cell proliferation.

Endocrinology 1988; 123:168-179.

66. May JV, Frost JP, Bridge AJ. Regulation of granulosa cell proliferation: facilitative

roles of platelet-derived growth factor and low density lipoprotein. Endocrinol-

ogy 1990; 126:2896-2905.

67. Skinner MK, Coffey Jr. RJ Regulation of ovarian cell growth through the local

production of transforming growth factor-a by theca cells. Endocrinology 1988;

123:2632-2638.

68. Adashi EY, Resnick CE, Hurwitz A, Ricciarelli E, Hernandez ER, Rosenfeld RG.

Ovarian granulosa cell-derived insulin-like growth factor binding proteins: mod-ulatory role of follicle-stimulating hormone. Endocrinology 1991; 128:754-760.

69. Bley MA, Simon JC, SaragUeta PE, Baranao JL. Hormonal regulation of rat gran-

ulosa cell deoxyribonucleic acid synthesis: effect of estrogens. Biol Reprod 1991;

44:880-888.

70. Knecht M, Feng P, Can K. Bifunctional role of transforming growth factor-Il dur-

ing granulosa cell development. Endocrinology 1987; 120:1243-1249.

71. MondscheinjS, Schomberg DW. Growth factors modulate gonadotropin receptorinduction in granuloss cell cultures. Science 1981; 211:1179-1180.

72. MondscheinJS, Smith SA, HammondJM. Production of insulin-like growth factor

binding proteins (IGFBPs) by porcine granulosa cells: identification of IGFBP-2

and -3 and regulation by hormones and growth factors. Endocrinology 1990;

127:2298-2306.

73. Mulheron GW, Schomberg DW. Rat granulosa cells express transforming growth

factor-Il type 2 messenger ribonucleic acid which is regulatable by follicle-stim-

ulating hormone in vitro. Endocrinology 1990; 126:1777-1779.74. Schomberg DW, May JV, Mondschein JS. Interactions between hormones and

growth factors in the regulation of granulosa cell differentiation in vitro. J Steroid

Biochem 1983; 19:291-295.75. Eppig B. Mouse oocyte development in vitro with various culture systems. Dcv

Biol 1977; 60:371-388.

76. Eppig 11, Schroeder AC. Capacity of mouse oocytes from preantral follicles to

undergo embryogenesis and development to live young after growth, maturationand fertilization in vitro. Biol Reprod 1989; 41:268-276.

77. Daniel SAJ, Armstrong DI, Gore-Langton RE. Growth and development of rat

oocytes in vitro. Gamete Rca 1989; 24:109-121.