BIOLOGY OF REPRODUCTION 54, 660-669 (1996)
Influence of Growth Factors on Proliferation and Morphogenesis
of Rabbit Ovarian Mesothelial Cells In Vitro1
Emilia Pierro,3 Santo V. Nicosia, 2 '4 Beatriz Saunders," Caroline B. Fultz,'
Roberto F. Nicosia, s and Salvatore Mancuso 3
Departmentof Obstetricsand Gynecology, 3 Catholic University of Sacred Heart, 00168 Rome, Italy
Departmentof Pathology,4 University of South Florida College of Medicine
and H. Lee Moffitt Cancer Center, Tampa, Florida 33612
Departmentof Pathology,5 Medical College of Pennsylvania, Philadelphia,Pennsylvania 19129
ABSTRACT
The ovarian mesothelium (OM) isthe source of one of the most frequent and lethal types of common ovarian epithelial tumors, the
so-called papillary serous carcinomas. Recent work from our laboratory indicates the existence of postovulatory luteal OM mitogens with
variable affinity for heparin. To further investigate the paracrine regulation of this ovarian tissue, rabbit ovarian mesothelial cells (OMC)
were cultured inserum-free, fibronectin-rich HL-1 medium with or without one of the following luteal growth factors (0.1, 1, and 10 ng/
ml): basic (bFGF) and acidic (aFGF) fibroblastic growth factors, epidermal growth factor (EGF), transforming growth factors a (TGFe) and
P(TGFP), tumor necrosis factor a, platelet-derived growth factor (PDGF-BB), and vascular endothelial growth factor. After 8 days of culture,
OMC growth was stimulated 3-fold by all tested doses of bFGF, EGF, and TGFa and 2-2.5-fold by 10 ng/ml of PDGF-BB and aFGF; it was
inhibited more than 60% by TGFP (10 ng/ml). Inaddition to enhancing the formation of cohesive OMC monolayers, most factors enhanced
3-to 6-fold the aggregation of OMC into papillary processes. The finding of a growth and morphogenetic response to intraluteal growth
factors is novel and suggests a role for postovulatory paracrine regulation of OM pathobiology.
INTRODUCTION
docrine, paracrine, or autocrine OMC pathways [17, 18]. It
has also been suggested that activation of such pathways
combined with exposure to environmental carcinogens
may lead to malignant transformation of OMC [19-21] and
to the development of the most lethal gynecological malignancy, known as ovarian epithelial cancer [3, 22]. In view
of the potential relevance of paracrine and autocrine cues,
the present study investigated the effect of a number of
growth factors present in CL [14, 23-26] on the growth and
morphogenesis of OM in serum-free culture.
The ovarian mesothelium (OM), commonly called surface
epithelium, undergoes morphologic changes throughout life
in response to physiologic and pathologic reproductive
events including ovulation [1], hormonal dysfunctions [2],
and pelvic injury [3, 4]. The occurrence of such changes in
the OM and not in adjacent extraovarian mesothelia suggests
that the OM is a modified mesothelium whose dynamic behavior may be influenced by intraovarian factors [5]. Previous
studies have demonstrated that ovarian mesothelial cells
(OMC) can express hormone receptors [6] and grow in response to steroid hormones [7] and gonadotropins [8]. Alternative growth signals for OMC may include paracrine stimuli
that derive from the underlying CL. For example, a marked
increase of DNA synthesis and morphogenesis is observed
in the postovulatory rabbit OM [5, 9]. Furthermore, CL tissue
extracts stimulate the growth of cultured rabbit OMC [10, 11].
Some studies also suggest that OMC may regulate their own
growth through autocrine mechanisms [12, 13], and the presence of mRNA for different growth factors has been demonstrated in normal rabbit OMC [14] as well as in normal and
neoplastic human OMC [15, 161.
On the basis of these studies, it has been postulated that
repair of the postovulatory defect may be regulated by en-
MATERIALS AND METHODS
Animals
Ovaries were obtained from 8 New Zealand white estrous rabbits (age: 4-6 mo). Animals were fed Purina chow
(Ralston-Purina Co., St. Louis, MO) ad libitum, individually
caged for a minimum of 3 wk, and killed by a pentobarbital
overdose according to principles and procedures outlined
in the NIH Guidelines for Care and Use of Experimental
Animals.
Chemicals
The dissociating enzymes used were Clostridium histolyticum collagenase type I (Sigma Chemical Company, St.
Louis, MO) and trypsin-EDTA (Gibco Laboratories, Grand Island, NY). BSA (fraction V) and plasma fibronectin were obtained from Sigma. The following tissue culture products
were obtained from Gibco: medium 199 (M199) with Hanks'
Accepted October 23, 1995.
Received May 30, 1995.
'This work was supported by VA Merit and NIH HL43392 grants.
'Correspondence: Santo V. Nicosia, MI)., I)epartment of Pathology MDC 11, University
of South Florida. College of Medicine, 12901 Bruce B. Downs Blvd., Tampa, FL 33612. FAX:
(813) 632-1708.
660
GROWTH FACTORS AND OVARIAN MESOTHELIAL CELLS
salts; Hanks' balanced salt solution (HBSS) with or without
Ca2 + and Mg 2+; mycoplasma-tested and complement-free fetal bovine serum (FBS); and L-glutamine. HL-1 medium was
purchased from Ventrex Laboratories (Portland, ME). This
chemically defined culture medium contains proprietary
amounts of insulin, transferrin, testosterone, sodium selenite,
ethanolamine, and HEPES buffer [8]. Cell isolation and culture media antibiotics consisted of penicillin and streptomycin (Gibco). Insulin, apo-transferrin, and glucose were obtained from Sigma. Bovine acidic fibroblast growth factor
(aFGF) and recombinant human basic fibroblast growth factor (bFGF) were purchased from Upstate Biotechnology
(Lake Placid, NY). Human recombinant platelet-derived
growth factor (PDGF-BB), epidermal growth factor (EGF),
transforming growth factor a (TGFa), and tumor necrosis factor a (TNFa) were obtained from Collaborative Biomedical
Products (Bedford, MA). Ultrapure human transforming
growth factor P, (TGF3) was purchased from Genzyme
(Cambridge, MA). Human recombinant vascular endothelial
growth factor (VEGF) was obtained from R&D Systems (Minneapolis, MN). The Cell Titer 96TM Aqueous Cell Proliferation assay was procured from Promega (Madison, WI).
Cell Isolation
OMC were isolated as previously described [27]. Briefly,
ovaries were removed under aseptic conditions, washed
with HBSS, and incubated for 1 h in collagenase (300 U/ml
of M199) at 37°C under a 5% CO2:95% air atmosphere. Ovaries were then resuspended in M199, vortexed for 60 sec,
and gently scraped with a no. 11 surgical blade under a
dissecting microscope. The cell-rich medium was placed in
a 5% BSA gradient for 20 min in order to separate OM fragments from single cells (red blood cells and fibroblasts). OM
fragments were then resuspended in Ca2 +/Mg2+-free HBSS
and further dissociated for 30 min in 0.05% trypsin/0.02%
EDTA at 37°C under a 5% CO2:95% air atmosphere. Cell
viability was determined by the trypan blue dye exclusion
test [27].
Cell Culture
Cells were seeded into multiwell plates (well diameter 2
cm; Becton Dickinson Labware, Rutherford, NJ) previously
coated with fibronectin (8 gg/ml of HL-1 medium) at a density of 5 x 103 cells per well (total medium volume: 0.25 ml).
Triplicate cultures were incubated for 8 days at 37°C in a 5%
CO2:95% air atmosphere. Cells were initially cultured for up
to 3 days in serum-free HL-1 medium supplemented with 4
gg/ml fibronectin, 4 mM L-glutamine, and 100 U/ml penicillin
and streptomycin. On Day 3 and Day 6, media were changed
and the growth factors (acidic and basic FGF, PDGF-BB,
VEGF, EGF, TGFa and 3, and TNFa) were added at concentrations of 0.1, 1, and 10 ng/ml. Human microvascular endothelial cells (HMEC) were utilized as positive controls for
661
angiogenic growth factors, particularly for VEGF, which has
been reported to have target specificity for endothelial cells
[24]. HMEC were propagated in T75 flasks and fed with M199
supplemented with 10% FBS and 20 U insulin, 2.5 mg apotransferrin, and 40 mg glucose/100 ml of culture medium.
On Day 0, HMEC were dispersed with 0.05% trypsin/0.02%
EDTA, washed, seeded at a density of 5000 cells in multiwell
plates, and cultured as described for OMC.
Growth Analysis
The cell titer assay used to evaluate cell growth contained
MTS (3-[4,5-dimethylthiazol-2-yl]-5-[3-carboxymethoxyphenazil]-2-[4-sulfophenil]-2H-tetrazolium, inner salt) and the
electron coupling reagent phenazine methosulfate (PMS).
MTS is bioreduced by cells into a water-soluble formazan
that can be quantitated at 490 nm. Optical density values
are linearly proportional to the number of living cells at a
concentration range of 6.25-50 X 103 cells as determined
in a standard assay. At the end of each culture, cells were
incubated with 333 pg/ml MTS and 25 lgM PMS for 2 h, and
optical density values were determined in a spectrophotometer [28, 29]. Experimental values were then divided by the
respective control values to obtain the percentage cell
growth.
Growth Characteristicsof OMC
Throughout culture, the growth pattern of OMC was
evaluated under phase-contrast microscopy using a Leitz
Labovert (Leica, Deerfield, IL) inverted microscope for assessment of the presence of epithelioid or fibroblastoid
morphologies as well as the number of papillary processes.
Such structures represent tridimensional cell aggregates that
rise abruptly from adjacent OMC monolayers [30] and display smooth and refractile borders like those of native OMC
papillae [31].
Immunocytochemistry
Ovaries were fixed in 10% neutral buffered formalin and
embedded in paraffin; 4-iim-thick sections were stained with
hematoxylin-eosin [32] to assess the effect of cell isolation
on the integrity of ovarian cortex. To evaluate the purity of
isolated cells and the growth characteristics of all cultures,
cells were fixed in 95% ethanol and immunostained to demonstrate low-molecular-weight (LMW) cytokeratin, vimentin, and factor VIII-related antigen (FVIII-RAg) using an
avidin-biotin complex (ABC) immunoperoxidase kit (Vectastatin Elite ABC; Vector Laboratories, Burlingame, CA) and
hematoxylin as nuclear counterstain [33]. Freshly isolated
OMC were spun at 1500 rpm for 5 min in a cytocentrifuge
(Shandon, Sewickley, PA), while cultured OMC were processed in situ within individual wells. Mouse monoclonal
anti-human cytokeratin antibody diluted 1:3 (Becton Dickinson, SanJose, CA), mouse monoclonal anti-human vimentin
diluted 1:200 (Biogenex Laboratories, San Ramon, CA), and
662
PIERRO ET AL.
FIG. 1. Immunocytochemistry of OMC (A-E) and human microvascular endothelial cells (F). Note diffuse expression of LMW cytokeratin (A)and vimentin (B)in OMC after
isolation. Note also diffuse cytokeratin expression in a trypsinized OMC culture (D)and nests of epithelioid, cytokeratin-positive OMC surrounded by tridimensional ridges
of smaller, oval- to spindle-shaped cytokeratin-positive OMC inan 8-day-old culture (E). In contrast with microvascular endothelial cells (F),OMC do not immunostain for
FVIII-RAg (C). A-C, x460; E, x290; F, x70.
FVIII-RAg diluted 1:25 [34] were used. Positive and negative
(nonimmune serum) controls were included in all immunoreactions.
Electron Microscopy
For transmission electron microscopy, OMC were fixed
in 2.5% glutaraldehyde in 0.1 M phosphate buffer and processed in situ [35]. Cells were postfixed in 1% osmium tetroxide, gradually dehydrated in alcohol, and embedded in
EMBed 812 (Electron Microscopy Systems, Fort Washington,
PA). Ultrathin sections (60-80 nm thick) were stained with
lead citrate and uranyl acetate and examined under a Philips
(Philips Electronics, Mahnah, NJ) 301 electron microscope
operated at initial accelerated voltage of 60 kV. For scanning
electron microscopy, 2.5% glutaraldehyde-fixed cells were
washed in 0.1 M phosphate buffer and further processed
after selected 0.5-1.0-cm 2 squares were sawed off from
each culture well [30]. Cells were osmicated, dehydrated
through graded alcohols, and substituted in the exchanging
medium hexamethyldisialazane [36]. Cultures were then
sputter-coated on a Hummer 5 (Anatech, Alexandria, VA)
device with a 15-20-nm-thick layer of gold-palladium
(60:40) and examined in a Jelco (JEOL, Peabody, MA) SM35 microscope operated at 10 kV with a stage tilt of 0-60° .
663
GROWTH FACTORS AND OVARIAN MESOTHELIAL CELLS
StatisticalAnalysis
TABLE 1. Effect of growth factors on OMC growth.'
Data were expressed as mean values
SD and examined by one-way ANOVA and by Duncan's Multiple Range
test for group homogeneity through use of a microcomputer-based statistical package (SPSS, Chicago, IL).
Growth factor
RESULTS
OMC Isolation
Before cell isolation, rabbit ovaries were covered by a continuous layer of cuboidal or low columnar, monostratified
OMC that occasionally formed papillae of various heights. A
thin basement membrane separated the papillary and nonpapillary OM from an underlying ovarian cortex containing
fibroblasts, microvessels, luteinized interstitial cells, and follicles of various developmental stages. As in previous studies
[231, exposure to collagenase followed by gentle surface
scraping separated the OM from the underlying basement
membrane and tunica albuginea. Unit gravity sedimentation
of isolated OM produced highly purified preparations of
OMC organoids equivalent after trypsinization to 3-6 X 105
OMC per ovary with a viability of 90-95%.
Characterizationof OMC
The purity of OMC was confirmed by universal and
strong LMW cytokeratin immunostaining (Fig. 1, A, D, and
E) and by a moderately strong and equally diffuse coexpression of vimentin (Fig. 1B). Vimentin immunostaining
was not observed in the absence of concomitant cytokeratin
expression, indicating the absence of contaminating fibroblasts. Isolated cells did not express FVIII-RAg (Fig. 1C),
indicating the absence of contaminating vascular endothelial cells. All cultured HMEC expressed FVIII-RAg (Fig. F).
Effect of Growth Factorson Formationof Papillary
Processes
In agreement with previous observations [8], OMC attached to fibronectin-coated dishes within 24-36 h of explantation. After 5-6 days, these cells formed mosaic-like
epithelioid monolayers (Fig. 2A) and focally organized into
nests and papillary processes (Fig. 2B). These structures
arose abruptly within OMC monolayers (Figs. 2C and 3A)
and were also observed at the leading edges of monolayers
in nonsaturated cultures. Most processes contained 10-20
cells, with constitutive cells displaying microvilli (Figs. 2D
and 3B), intermediate cell junctions (Fig. 3C), and frequent
cisternae of granular endoplasmic reticulum surrounding
electron-dense material (Fig. 3D). OMC deep within papillary processes were occasionally separated by spaces containing a fibrillary matrix.
The number of papillary processes markedly increased
when cells were cultured in the presence of some of the
aFGF
bFGF
PDGF
VEGF
Control 2
EGF
TGFa
TGFP
TNFa
Control 2
0.1 ng/ml
0.046
0.07
0.041
0.046
+ 0.017
± 0.028*
+ 0.023
+ 0.022
0.201
0.222
0.115
0.11
+ 0.076**
± 0.065**
± 0.01
+ 0.031
1 ng/ml
0.055
0.089
0.058
0.052
0.035
0.27
0.268
0.11
0.105
0.107
±+0.021
+ 0.033**
+ 0.026
± 0.024
± 0.018
± 0.083**
± 0.049**
+ 0.029
+ 0.027
+ 0.029
10 ng/ml
0.084
0.102
0.09
0.058
± 0.035**
0.305
0.307
0.054
0.124
± 0.1**
± 0.073**
+ 0.029*
± 0.02
± 0.031**
- 0.035*
+ 0.031
'Values are indicated as optical density at 490 nm. *p < 0.05; **p < 0.01.
Four growth factors and controls evaluated intriplicate batches.
2
growth factors tested (Fig. 4). The highest increments were
observed in cultures stimulated with bFGF (640%, 790%,
and 1000% at 0.1, 1, and 10 ng/ml, respectively), EGF
(570%, 670%, and 780% at 0.1, 1, and 10 ng/ml, respectively), and TGFa (550%, 730%, and 750% at 0.1, 1, and
10 ng/ml, respectively). Significant differences were also
observed in the presence of 1 and 10 ng/ml of aFGF
(380% and 620%, respectively), 1 and 10 ng/ml of PDGFBB (500% and 690%, respectively), and 1 and 10 ng/ml of
VEGF (420% and 440%, respectively). In contrast, TGF[ at
10 ng/ml reduced the aggregation of OMC into papillary
processes to half of control values.
Effect of Growth Factorson OMC Proliferation
Cultures were exposed to three different concentrations
(0.1, 1, and 10 ng/ml) of each growth factor (Table 1). At
Day 7, significant enhancement of cell growth was noted
after treatment with aFGF, bFGF, PDG-BB, EGF, and TGFct.
The most pronounced effect was seen in the presence of 10
ng/ml bFGF (315
83%), EGF (278
40%), and TGFa
(288 + 17%) (Fig. 5). These three growth factors also elicited significant dose-dependent stimulation of cell proliferation at lower, i.e., 0.1 and 1.0 ng/ml, concentrations (209
+ 45% and 273 + 74% for bFGF; 183 + 28% and 248 +
32% for EGF; and 198
29% and 248 + 21% for TGFt,
respectively) (Fig. 5).
FIG. 2. (page 664) Growth characteristics of OMC after 8 days in serum-free, fibronectin-rich medium. Note epithelioid OMC monolayer (A) and tridimensional
OMC aggregates (B,arrows). The papillary-like morphology and microvillar surface
of these aggregates are better highlighted under scanning electron microscopy (CD). A-B, x 255; C, x 1175; D, x4600.
FIG. 3. (page 665) Transmission electron microscopy views of OMC after 8 days
in serum-free, fibronectin-rich medium. A) Note transition between monolayered
OMC and stratified OMC of a papillary-like aggregate (arrow). Culture vessel substrate is indicated by the asterisk. B) Core of papillary aggregate displays multiple
layers of OMC that focally surround an extracellular lumen (asterisk); microvilli are
noted in OMC lining the lumen as well on as the apical surface of the aggregate
(arrow). OMC are joined by intermediate junctional devices (C,arrow) and contain
numerous free ribosomes (D,arrow) and focally dilated (D, asterisks) cisternae of
granular endoplasmic reticulum. A, x 1372; B, x 2548; C-D, x 20 090.
664
PIERRO ET AL.
GROWTH FACTORS AND OVARIAN MESOTHELIAL CELLS
665
666
PIERRO ET AL.
CONTROL
CONTROL
aFGF
-~~-m~~*
aFGF
bFGF
*
--
o10.1 ng/ml
PDGF
-- **~~
VEGF
I1
1.0 ng
0O ng
EGF.O
__
TGF
TNF
.,.
iO.1 ng
...................
.
VEGF
'--**
TGF_
**
PDGF
ng/ml
E10 ng/ml
EGF,
*
bFGF
**
TGF
_
0
2
4
6
10
8
**
TGF
12
TNF
# PAPILLARY PROCESSES
0
FIG. 4. Number of OM papillary processes in the presence of various growth fac-
50
100
150
200
250
300
350
%GROWTH
tors. * p <0.05. **p <0.01.
FIG. 5. Influence of growth factors on OMC growth. Effects are indicated as percentage of growth relative to controls. * p < 0.05. ** p < 0.01.
Statistically significant stimulation of cell growth was also
observed after treatment with the highest concentrations (10
ng/ml) of aFGF (217 ± 35%) and PDGF-BB (258 ± 5.28%)
(Table 1; Fig. 5). No significant effect was seen with lower
concentrations of either of these growth factors. At the dose
of 10 ng/ml, TGFP markedly inhibited cell growth (56 +
11.9%) (Fig. 5). No significant changes were observed after
treatment with any dose of VEGF and TNFa.
All growth factors (aFGF, bFGF, PDGF, VEGF, EGF, and
TGFa) known to be mitogenic for endothelial cells in culture [23, 24] stimulated the growth of HMEC at all tested
doses (Table 2). Control and growth factor-exposed OMC
cultures coexpressed cytokeratin (Fig. 1E) and vimentin and
did not stain with FVIII-RAg, thus ruling out the emergence
of cell phenotypes other than OMC.
DISCUSSION
The present study was designed to evaluate the effect of
a number of growth factors, previously demonstrated in
rabbit CL [10, 11, 14, 23-26, 37], on cultured OMC. The results can be summarized as follows: 1) OMC growth was
enhanced by growth factors with high (aFGF, bFGF), low
(PDGF, EGF), and little or no (TGFa) affinity for heparin; 2)
TGF3 inhibited OMC growth while TNFa had no measurable effects; 3) OMC developed in vitro papillary processes,
and the morphogenesis of such structures was maximally
stimulated by bFGF, EGF, and TGFa.
As for other cell types [38, 39], proliferative effects of
growth factors have been reported for normal and neoplastic ovarian [40, 41] and extraovarian [42] mesothelial cells.
The existence of mesotheliogenic growth factors (including
FGFs, TGFa, and EGF) has been demonstrated in rabbit CL,
ovarian interstitial cells, and OMC by reverse transcriptionpolymerase chain reaction and Western immunoblotting
[14, 23-26]. In situ immunocytochemical and image analysis studies have also shown quantitative temporal changes
in growth factors during development and aging of CL, with
the highest cytoplasmic expression at Day 12 after ovulation
[14]. These temporal changes correlate with variations in the
growth-stimulating and morphogenetic activities of crude
CL extracts on cultured OMC [10]. All such findings suggest
that OMC growth signals may originate from underlying folliculo-luteal complexes and that such signals may influence
the growth of OMC after rupture of the follicle [1, 5]. Proliferation of OMC plays an important role in repairing the tissue defect caused by ovulation, and such an event indeed
makes the OMC one of the most proliferative, albeit focally
and cyclically, ovarian constituents [9, 37].
TABLE 2. Effect of growth factors on HMEC.'
Growth factor 2
aFGF
bFGF
PDGF
VEGF
EGF
TGFa
Control
1
0.1 ng/ml
0.587
0.846
0.461
0.565
0.558
0.612
+ 0.055 (170
± 0.118 (241
± 0.031 (133
± 0.061 (163
+ 0.086 (183
± 0.064 (192
1 ng/ml
±
±
±
±
±
±
18.3)
6.3)
4.9)
1.4)
24.8)
29.5)
0.684
0.893
0.69
0.691
0.565
0.76
± 0.11 (198 ±
± 0.091 (263 ±
± 0.087 (200 ±
± 0.076 (201 ±
± 0.229 (218 ±
± 0.057 (240 ±
0.329 ± 0.056
10 ng/ml
14.8)
31.1)
26.1)
2.8)
24.8)
19)
0.761
0.667
0.908
0.616
0.804
0.737
± 0.058(221
± 0.037 (194
± 0.028 (263
± 0.061 (179
± 0.056 (252
+ 0.256 (264
± 4.2)
± 9.1)
± 0.7)
+ 0)
± 24.9)
± 19.6)
Values are indicated as optical density at 490 nm. Numbers in parentheses are percentage change relative to controls. All points
showed a significant difference compared to controls (p < 0.05).
2
Controls and growth factors evaluated in triplicate batches.
GROWTH FACTORS AND OVARIAN MESOTHELIAL CELLS
As suggested by the present study, the mechanisms that
regulate postovulatory repair are poorly understood but
might involve the activation of paracrine pathways. Such
pathways may depend on the cellular secretion of soluble
growth factors or on their release from extracellular matrix
(ECM) [43]. Autocrine mechanisms may also be operative,
since constitutive production of macrophage colony-stimulating factor, interleukin 6, and other cytokines has been
demonstrated in human OMC [12, 40, 41]. Because of their
angiogenic activity [23, 44], some luteal growth factors may
also play a role not only in CL development but also in the
formation of the vascular axis of native OM papillae. This
potential angioformative activity is supported by the strong
in vitro angiogenic effect of CL extracts (unpublished results) and purified growth factors [45]. A better understanding of the molecular regulation of postovulatory repair may
be important in view of the relationship between repetitive
ovulations and OM cancer [21, 46, 47]. Such a relationship
remains to be further clarified in view of the increased incidence of OM cancer postmenopausally, when ovulation
and CL formation are absent or rare [18]. As for other cancers, it is possible that malignant transformation may result
from cumulative postovulatory growth pressure events
leading to cytogenetic instability of OMC [47].
In this study, the highest enhancement of OMC growth
was observed after treatment with EGF, TGFa, and bFGF.
Of these growth factors, bFGF is the one with the highest
affinity for the ECM [23]. Unlike EGF, TGFa, and PDGF-BB,
which are released in soluble forms, bFGF lacks a signal
sequence and binds to the heparan sulfate proteoglycans of
ECM [43]. OMC are capable of producing some ECM components autonomously [30, 48] or under the influence of
growth factors [49], thus regulating their growth and ability
to organize into papillary processes in vivo and in vitro
[30, 48, 50]. Although we have noted the formation of OMC
papillae in nonconfluent monolayers, continuous availability of ECM may allow OMC to circumvent the restraints of
density-dependent growth inhibition in vitro [48]. In addition, OMC-generated ECM may provide a reservoir of
paracrine and autocrine growth factors necessary for OMC
growth and papillogenesis in vitro and in vivo after ovulation or during neoplastic progression [37, 48, 49]. OMC can
physically modulate the ECM directly or under the influence
of TGF[3 and PDGF [51], and such properties may facilitate
the translocation of these cells over developing papillae.
Noteworthy in the present study is the similarity between
the effects of EGF and TGFa on OMC. The finding is consistent with previous studies showing that these growth factors are almost equivalent in their ability to stimulate DNA
synthesis in various cell lines [38] and to compete for the
EGF receptor [52] known to be expressed in both normal
and neoplastic OMC [15, 53]. TGFB3, unlike the other tested
growth factors, is a bifunctional growth-regulatory peptide
[54]. In general, this factor inhibits epithelial cell growth
667
while it may either stimulate or inhibit the growth of mesenchymal cells [54]. In the present and other [55] work,
TGF3 markedly inhibited the growth of normal and neoplastic OMC. It is known that normal OMC can produce
TGF,3 and TGF32 [40] and that some ovarian cancer cell
lines can produce TGF3 [41]. It is thus possible that loss of
the TGFP autocrine inhibitory pathway may be involved in
unregulated growth and eventual neoplastic progression of
OMC [411.
Papillae are tridimensional processes composed of variable layers of OMC resting on a basement membrane and
a fibrovascular axis [31, 48]. In the ovaries of some mammals, including the rabbit [50], OMC stratification and papillae are frequent, while the OM is usually monostratified
in the normal human ovary except during fetal life [56].
However, prominent papillae have been observed in association with some human pathological conditions such as
the polycystic ovary and luteinized unruptured follicle syndrome [2]. The first condition is characterized by elevation
in androstenedione or testosterone and estradiol and by
changes in gonadotropins, while the second is associated
with elevation in progesterone levels not accompanied by
ovulation. FSH and LH as well as chorionic gonadotropins
stimulate to various degrees the proliferation of OMC in vitro [8], and estradiol is highly papillogenic in vivo [37]. Morphogenesis of OM papillae is also exuberant in rabbits after
ovulation [5, 9] and in humans after pathologic injuries
caused by pelvic inflammations or endometriosis [4]. Papillae in addition are characteristic components of low-malignant and frankly malignant ovarian serous neoplasms [3].
The finding that some luteal growth factors stimulate not
only OMC growth but also aggregation into papillary processes suggests that these factors may play a major role in
OM morphogenesis under physiologic and pathologic conditions [57]. In our study, enhancement of OMC aggregation
was also noted with higher doses of VEGF. Such effect was
unexpected because of the reported specificity of this factor
for endothelial cells [23, 44]. In view of the diffuse cytokeratin positivity and a lack of endothelial antigen immunoreactivity in cultured OMC, this finding needs to be investigated further.
The findings of the present study are of potential significance. On the basis of the involvement of oncogenes and
oncogene products in carcinogenesis [38-41], some of the
growth factors evaluated in the study may be involved in the
endogenous regulation of OMC leading to dysregulated
growth and cancer. If indeed OMC growth and morphogenesis are growth factor-dependent, an array of autocrine and
paracrine interactions may occur among OMC and between
these cells and adjacent ovarian cells. Endocrine interactions
may also take place, since growth factors such as EGF can
modulate gonadotropin receptors [58] and reproductive tract
growth and differentiation [59]. It is likely that the consequences of growth factor-receptor interactions with specific
668
PIERRO ET AL.
cell types would be diverse and complex. In order to better
understand such interactions, an in vitro model has been recently developed in our laboratory [60]. In such a model,
humoral and cellular interactions can be ideally studied to
assess the significance of autocrine and paracrine regulation
of OMC growth and papillogenesis. We believe that an increasing understanding of such important events may lead
to better insight into fundamental physiologic and pathologic
OM events, including the initiation of cancer.
ACKNOWLEDGMENTS
The authors thank Dr. Siamak Tabibzadeh for the generous gift of HMEC. The technical
contributions of Ms. Terri Stella-Vega and Mr. Edward Haller and the secretarial assistance
of Mrs. Joyce Campbell and Judy Wasserberger are also gratefully acknowledged.
REFERENCES
1. Motta P, Van Blerkom J, Makabe S. Changes in the surface morphology of ovarian germinal epithelium during the reproductive cycle and in some pathological conditions. A
correlative three dimensional analysis by transmission, scanning and high voltage elec
tron microscopy. J Submicrosc Cytol 1980; 407-425.
2. Makabe S, Iwaki A, Hafez ESE, Motta PM. Physiomorphology of fertile and infertile
human ovaries. In: Motta PM, Hafez ESE (eds.), Biology of the Ovary. Boston: Martinus
Nijhoff; 1980: 280-290.
3. Nicosia SV, Nicosia RF. Neoplasm of the ovarian mesothelium. In: AzarHA (ed.), Pathology of Human Neoplasms. New York: Raven Press; 1988: 435-486.
4. Farhi DC, Silvenberg SG. Pseudometastases in female genital cancer. Pathol Annu 1982:
17:47-76.
5. Osterholzer HO, Johnson JH, Nicosia SV. An autoradiographic study of rabbit ovarian
surface epithelium before and after ovulation. Biol Reprod 1985; 33:729-738.
6. Hamilton TC, Henderson WJ, Eaton C. Steroid hormone receptor status of the normal
and neoplastic ovarian surface germinal epithelium. In: Greenwald GS, Terranova PF
(eds.), Factors Regulating Ovarian Function. New York: Raven Press; 1983: 81-85.
7. Hamilton TC, Henderson WJ, Eaton C. Isolation and growth of the rat ovarian germinal
epithelium. In: Richards RJ,Rajan KT (eds.), Tissue Culture in Medical Research. Oxford:
Pergamon; 1980: 237-244.
8. Osterholzer HO, Streibel EJ, Nicosia SV. Growth effects of protein hormones on cultured
rabbit ovarian surface epithelial cells. Biol Reprod 1985; 33:247-258.
9. Nicosia SV, Oliveros-Saunders B, Narconis RJ. Regulation and temporal sequence of
surface epithelium morphogenesis in the postovulatory rabbit ovary. Prog Clin Biol Res
1989; 296:111-119.
10. Nicosia SV, Oliveros-Saunders B. Initial characterization of a luteal growth factor for
ovarian mesothelial cells. In: Hirshfield AN (ed.), Growth Factors and the Ovary. New
York: Plenum Publishers; 1989: 237-244.
11. Setrakian S, Oliveros-Saunders B, Nicosia SV. Growth stimulation of ovarian and extraovarian mesothelial cells by corpus luteum extract. In Vitro Cell Dev Biol 1993; 879883.
12. Lidor YJ, Xu GJ, Martinez-MazaO, Olt GJ, Marks JR, Berchuck A,Ramakrishnan S, Berek
JS, Bast RC Jr. Constitutive production of macrophage colony-stimulating factor and interleukin-6 by human ovarian surface epithelial cells. Exp Cell Res 1993; 207:332-339.
13. Berchuck A, Rodriguez GC, Olt GJ, Boente MP, Whitaker R, Arrick B. Regulation of
growth of normal ovarian epithelial cells and ovarian cancer cell lines by transforming
growth factor-P. AmJ Obstet Gynecol 1992; 166:676-684.
14. Nicosia SV, Oliveros-Saunders B, Sun X, Bouillard C, Fultz C, Giacomini G, Jasonni V.
Expression and modulation of growth factors in rabbit corpora lutea. Mol Biol Cell 1993;
4:442.
15. Rodriguez GC, Berchuck A, Whitaker RS, Shlossman D, Clarke-Pearson DL, Bast RCJr.
Epidermal growth factor receptor expression in normal ovarian epithelium and ovarian
cancer. II. Relationship between receptor expression and response to epidermal growth
factor. AmJ Obstet Gynecol 1991; 164:745-750.
16. Owens OJ, Stewart C, Leake RE. Growth factors in ovarian cancer. Br J Cancer 1991:
64:1177-1181.
17. Duke KL. Nonfollicular ovarian components. In: Jones RE (ed.), The Vertebrate Ovary.
Comparative Biology and Evolution. New York: Plenum Publishers; 1978: 563-582.
18. Nicosia SV. Morphological changes of the human ovary throughout life. In: Serra GB
(ed.), The Ovary. New York: Raven Press; 1983: 57-81.
19. GrahamJ, Graham R. Ovarian cancer and asbestos. Environ Res 1967; 1:115-128.
20. Hamilton TC, Fox H, Buckley CH, Henderson WJ, Griffiths K. Effects of talc on the rat
ovary. BrJ Exp Pathol 1984; 65:115-128.
21. Woodruff JD. The pathogenesis of ovarian neoplasia. Johns Hopkins Med J 1979;
144:117-120.
22. Blaustein A. Surface (germinal) epithelium and related ovarian neoplasm. Pathol Annu
1981; 16:247-294.
23. Koos RD. Potential relevance of angiogenic factors to ovarian physiology. Sem Reprod
Endocrinol 1989; 7:29-40.
24. Ferrara N, Houck KA, Jakeman LB, WinerJ, Leung DW. The vascular endothelial family
of polypeptides. J Cell Biochem 1991; 47:211-218.
25. Sun X, Oliveros Saunders B, Zhou D, Fultz C, Giacomini G, Jasonni V, Nicosia SV. Expression of basic fibroblast growth factor in mesothelial and extramesothelial compartments of the rabbit ovary. Mol Biol Cell 1993; 4:18.
26. Reynolds LP,Grazul-Bilska AT, Redmer DA. Mitogenic factors of corpora lutea. Prog
Growth Factor Res 1994; 5:159-175.
27. Nicosia SV,Johnson JH, Streibel EJ. Isolation and ultrastructure of rabbit ovarian mesothelium (surface epithelium). IntJ Gynecol Pathol 1984; 3:348-360.
28. Cory AH, Owen TC, Baltrop JA, Cory JG. Use of an aqueous soluble tetrazolium/formazan assay for cell growth assays in culture. Cancer Comm 1991; 3:207-212.
29. Scudiero DA, Shoemaker RH, Paull KD, Tierney S, Nofziger TH, Currens MJ, Seniff D,
Boyd MR. Evaluation of a soluble tetrazolium/formazan assay for cell growth and drug
sensitivity in culture using human and other tumor cell lines. Cancer Res 1988; 48:48274833.
30. Nicosia SV,Johnson JH, Streibel EJ. Growth characteristics of rabbit ovarian mesothelial
(surface epithelial) cells. IntJ Gynecol Pathol 1985; 4:58-74.
31. Cherney DD, Motta P, DidioLJ. Ovarian villi in rabbits studied with light, scanning and
transmission electron microscopy. J Microsc 1973; 17:37-40.
32. Luna LG. Manual of Histologic Staining Methods of the Armed Forces Institute of Pathology, 3rd ed. New York: McGraw-Hill; 1968: 38-39.
33. Hsu S, Raine L, Fanger H. The use of avidin-biotin-peroxidase techniques. J Histochem
Cytochem 1981; 29:577-580.
34. Czernobilsky B. Co-expression of cytokeratin and vimentin filaments in mesothelial.
granulosa and rete ovarii cells of the human ovary.EurJ Cell Biol 1985 37:175-190.
35. Kuhn H. A simple method for the preparation of cell cultures for ultrastructural investigation. J Histochem Cytochem 1981; 29:84-86.
36. BrayDF, Bagu J, Koegler P. Comparison of hexamethyldisilazane, Peldri II, and criticalpoint drying methods for scanning electron microscopy of biological specimens. Microsc
Res Techn 1993; 26:489-495.
37. Nicosia SV, Oliveros-Saunders B, Acedevo-Duncan ME, Setrakian S, De Gregorio R. Biopathology of ovarian mesothelium. In: Familiari G, Makabe S, Motta PM (eds.). Ultrastructure of the Ovary. Boston: Kluwer Academic Publ.; 1991: 287-310.
38. Deuel TF. Polypeptide growth factors: roles in normal and abnormal cell growth. Annu
Rev Cell Biol 1987; 3:443-492.
39. Goustin AS, Leof EB, Shipley GD, Moses HL. Growth factors and cancer. Cancer Res
1986; 46:1015-1029.
40. Berchuck A, OltGJ, Lorri Everitt BS, Soisson AP,Bast RCJr, Boyer CM. The role of peptide
growth factors in epithelial ovarian cancer. Obstet Gynecol 1990; 75:255-262.
41. Berchuck A, Kohler MF, Boente MP, Rodriguez GC, Whitaker RS, Bast RC Jr. Growth
regulation and transformation of ovarian epithelium. Cancer Suppl 1992; 71:545-551.
42. Breborowicz A,Rodela H, PagiamtzisJ, Oreopolus DG. Stimulation of mesothelial cells
proliferation by endogenous growth factors. Pent DyalInt 1991;11:228-232.
43. Vlodavsky I, Fridman R, Sullivan R, Sasse J, Klagsbrun M. Aortic endothelial cells syn
thesize basic fibroblast growth factor which remains cell associated and platelet-derived
growth factor-like protein which is secreted. J Cell Physiol 1987; 131:402-408.
44. D'Amore PA, Smith SR. Growth factor effect on cells of the vascular wall: a survey.
Growth Factors 1993; 8:61-75.
45. Nicosia RF, Nicosia SV, Smith M. Vascular endothelial growth factor, platelet derived
growth factor and insulin-like growth factor-I promote rat aortic angiogenesis. Am J
Pathol 1994; 145:1023-1029.
46. Fathalla MF Incessant ovulation-a factor in ovarian neoplasia? Lancet 2 1971 163169.
47. Godwin AK, Testa JR, Handel LM,Liu Z, Vanderveer LA,Tracey PA, Hamilton TC. Spontaneous transformation of rat ovarian surface epithelial cells: association with cytogenetic
changes and implications of repeated ovulation in the etiology of ovarian cancer. J Int
Cancer Inst 1992; 84:592-601.
48. Auersperg N,Maclaren IA, Kruk PA. Ovarian surface epithelium: autonomous production
of connective tissue-type extracellular matrix. Biol Reprod 1991; 44:717-724.
49. Gospodarowicz D, Lui GH. Effect of substrate and growth factors on the proliferation
in vitro of bovine aortic endothelial cells. J Cell Physiol 1981; 109:69-81.
GROWTH FACTORS AND OVARIAN MESOTHELIAL CELLS
50. Nicosia SV, Johnson JH. Surface morphology of ovarian mesothelium (surface epithelium) and of other pelvic and extrapelvic mesothelial sites in the rabbit. Int J Gynecol
Pathol 1984; 3:249-260.
51. Kruk PA, Auersperg N. Human ovarian surface epithelial cells are capable of physically
restructuring extracellular matrix. Am J Obstet Gynecol 1992; 167:1438-1443.
52. Massague'J. Epidermal growth factor-like transforming growth factor. II. Interaction with
epidermal growth factor receptors in human placenta membranes and A-431 cells. J Biol
Chem 1983; 258:13614-13620.
53. Henzen-Logmans SC, van der Burg MEL, Foekens JA, Berns PMJJ, Brusse' R, Fieret JH,
Klijn JGM, Chadha S, Rodenburg CJ. Occurrence of epidermal growth factors receptors
in benign and malignant ovarian tumours and normal ovarian tissues: an immunohistochemical study. J Cancer Res Clin Oncol 1992; 118:303-307.
54. Moses HL, Yang EY, Pietenpol JA. TGF-{3 stimulation and inhibition of cell proliferation:
new mechanism insights. Cell 1990; 63:245-247.
669
55. Hurteau J, Rodriguez GC, Whitaker RS, Shah S, Mills G, Bast RC, Berchuck A. Transforming growth factor-n inhibits proliferation of human ovarian cancer cells obtained
from ascites. Cancer 1994; 74:93-99.
56. Gondos B. Surface epithelium of the developing ovary. AmJ Pathol 1975; 81:303-320.
57. Nicosia SV, Nicosia RF, Saunders BO, Sun X, Pierro E, Mancuso S. Control of papillogenesis in benign and malignant ovarian mesothelium. In: Fujimoto S, Hsueh AJW,
Strauss JF, Tanaka T (eds.), New Achievements in Research of Ovarian Function. Frontiers in Endocrinology, Ares-Serono 1995:333-343.
58. Mondschein JS, Schomberg DW. Growth factors modulate gonadotropin receptor induction in granulosa cell cultures. Science 1981; 211:1179-1180.
59. Gupta C, Siegel S, Ellis D. The role of EGF in testosterone-induced reproductive tract
differentiation. Dev Biol 1991; 146:106-116.
60. Giacomini G, Nicosia SV,Saunders BO, Fultz C, Sun X,Jasonni VM. Ovarian mesothelial
and extramesothelial cells in interactive culture. In Vitro Cell Dev Biol 1995; 31:300-309.