Cytokine 43 (2008) 209–214

Contents lists available at ScienceDirect

Cytokine

journal homepage: www.elsevier .com/locate / issn/10434666

Immunohistochemical detection of insulin-like growth factor-I, transforminggrowth factor-b2, basic fibroblast growth factor and epidermal growthfactor-receptor expression in developing rat ovary

K. Ergin a,*, E. Gürsoy a, Yücel Bas�ımoglu Koca c, H. Bas�aloglu b, K. Seyrek d

a Department of Histology and Embryology, Adnan Menderes University School of Medicine, TR-09100 Aydın, Turkeyb Department of Anatomy, Adnan Menderes University School of Medicine, TR-09100 Aydın, Turkeyc Department of Biology, Adnan Menderes University Science and Arts, TR-09100 Aydın, Turkeyd Department of Biochemistry, Adnan Menderes University School of Veterinary Medicine, TR-09100 Aydın, Turkey

a r t i c l e i n f o a b s t r a c t

Article history:Received 6 June 2007Received in revised form 14 April 2008Accepted 14 May 2008

Keywords:Growth factorFollicleOvaryImmunohistochemistry

1043-4666/$ - see front matter � 2008 Elsevier Ltd. Adoi:10.1016/j.cyto.2008.05.013

* Corresponding author. Fax: +90 256 2123169.E-mail address: kemergin@yahoo.com (K. Ergin).

The aim of this study was to determine the immunohistochemical expression and localization of insulin-like growth factor-I (IGF-I), transforming growth factor-b2 (TGF-b2), basic fibroblast growth factor (bFGF)and epidermal growth factor-receptor (EGF-R) in developing rat ovaries.Eighteen female Wistar rats were enrolled in this study; newborn (n = 6), one-month-old (n = 6) and adult(n = 6) rats. Formalin-fixed and parafin-embedded ovarian tissues were stained with antibodies againstIGF-I, TGF-b2, bFGF and EGF-R, immunohistochemically. The ovarian cells were evaluated by semi-quan-titative scoring system under light microscope.The staining of IGF-I, TGF-b2, bFGF and EGF-R were most intense in the oocytes and were heavily at one-month-old rats. A moderate immunostaining in theca cells and corpus luteii reacted with IGF-I in adultrats. Furthermore the staining intensity for IGF-I was moderate in granulosa cells of newborn rat ovaries.We detected also a moderate staining for TGF-b2 in corpus luteii of adult rats. In addition, we found abFGF immunostaining mainly in oocytes of follicles of young and adult rats. Immunostaining for EGF-Rwas moderate in granulosa cells of one-month-old rats.In conclusion, this study suggests that growth factors play a pivotal role in ovarian function, especially infollicular development. The role of growth factor in controlling degeneration or growth (or both) of ovaryfollicles remain as explained.

� 2008 Elsevier Ltd. All rights reserved.

1. Introduction

More than 99% of follicles in ovary undergo atresia, during thereproductive life of the females [1]. Atresia occurs by apoptosis ofthe follicular cells and of the oocytes [2]. Successful follicle develop-ment depends on the presence of survival factors that protect cellsfrom apoptosis and also promote follicle growth in ovary [2]. Evi-dence suggest that growth factors influence ovarian physiology indifferent stages of follicular growth via endocrine, paracrine andautocrine mechanisms [3,4]. Growth factors seem to modulate cellsurvival and cell death balance between the granulosa cells [5] andplay an important role in the regulation of ovarian follicular develop-ment [2,6,7].

It was shown in several mammalian species including rats thatinsulin-like growth factor-I (IGF-I) stimulates both proliferationand differentiation of granulosa cells [2,6,7] and promotes granu-

ll rights reserved.

losa cell survival by increasing resistance to apoptosis [7]. IGF-I,basic fibroblast growth factor (bFGF) and epidermal growth factor(EGF) seem to have an important anti-apoptotic effect in ovariantissues [5,8,9]. It was also reported that IGF-I, bFGF and EGF sup-port growth of small antral follicles by enhancing granulosa cellproliferation [10]. IGF-I, EGF and transforming growth factor-beta(TGF-b) have also been shown to be involved in ovarian folliculardynamics [4,11]. It is also believed that bFGF have an importantrole in early folliculogenesis and can regulate primordial develop-ment [12,13]. In addition, it has been shown that TGF-b was crucialfor embryogenesis and for regulation of cellular activities such asproliferation and apoptosis, which are important for tissue growthand morphogenesis [14,15].

Multiple studies have been performed in various mammalianspecies to demonstrate the roles for IGF-I, TGF-b, bFGF and EGF-Rproteins in various aspects of follicular growth and ovarian deve-lopment. However, to our knowledge, there is no study that com-pares the expression and localization of IGF-I, TGF-b2, bFGF andEGF-R proteins in ovaries of newborn, one-month-old and adultrats. Thus, the aim of this study was to determine the expression

Table 2Expression and distribution of IGF-I, TGF-b2, bFGF and EGF-R in the stroma and

210 K. Ergin et al. / Cytokine 43 (2008) 209–214

and localization of IGF-I, TGF-b2, bFGF and EGF-R in the stage ofdeveloping rat ovary.

Table 1Expression and distribution of insulin-like growth factor-I (IGF-I), transforminggrowth factor-beta 2 (TGF-b2), fibroblast growth factor-2 (FGF-2) and epidermalgrowth factor receptor (EGF-R) in the follicles of developing rat ovary

Primordialfollicles

Primaryfollicles

Secondaryfollicles

Tertiaryfollicles

IGF-INewbornGranulosa

cells++ ++ Not available Not available

Oocyte � �Theca cells � �One-month-oldGranulosa

cells� + � Not available

Oocyte +++ +++ +++Theca cells � � �AdultGranulosa

cells� � � �

Oocyte � + ++ +Theca cells � � ++ ++

TGF-b2NewbornGranulosa

cells� � Not available Not available

Oocyte � �Theca cells � �One-month-oldGranulosa

cells� � � Not available

Oocyte + + +++Theca cells � � +AdultGranulosa

cells� � � �

Oocyte � ++ ++ �Theca cells � � � �

bFGF (FGF-2)NewbornGranulosa

cells� � Not available Not available

Oocyte � �Theca cells � �One-month-oldGranulosa

cells� � � Not available

Oocyte � +++ +++Theca cells � � �AdultGranulosa

cells� � � �

Oocyte � � ++ ++Theca cells � � � �

EGF-RNewbornGranulosa

cells� � Not available Not available

Oocyte +++ +++Theca cells � �One-month-oldGranulosa

cells+ ++ ++ Not available

Oocyte + ++ ++Theca cells � � �AdultGranulosa

cells� + + +

Oocyte � + + +Theca cells � � � �

The intensity of staining is grouped into categories: �, no staining; +, weak staining;++, moderate staining; +++, strong staining.

corpus luteum of developing rat ovary

IGF-I TGF-b2 bFGF EGF-R

StromaNewborn � � � �One-month- old +++ ++ + �Adult +++ ++ � +

Corpus luteumAdult ++ ++ � +

The intensity of staining is grouped into categories: �, no staining; +, weak staining;++, medium staining; +++, strong staining.

2. Material and methods

2.1. Animals and treatment

Newborn (n = 6), one monthly (n = 6) and adult (n = 6) femaleWistar rats were used in this study. They were kept at a constanttemperature (21 ± 1 �C) and controlled light conditions (light07.00–19.00). Food (standard pellet diet) and tap water were sup-plied ad libitum. Rats were anesthetized with ether and were killedby cervical dislocation for collection of ovary. All studies with ani-mals described here in, were reviewed and approved by the Uni-versity of Adnan Menderes Animal Ethics committee.

2.2. Immunolocalization

Ovary samples were fixed in 10% neutral buffered formalin, rou-tinely processed (dehydration steps in ethanol and clearing in xy-lene) and embedded in paraffin blocks. Tissues in paraffin blockswere randomly cut in 5 lm sections by a microtome (Leica RM2135) and placed on poly-L-lysine coated glass slides. Immunohis-tochemical staining was performed by avidin–biotin immunoper-oxidase system. After 2 h incubation at 40 �C, sections weredeparaffinized in xylene, hydrated through graded ethanol andendogenous peroxidase blocked with 3% H2O2 in 70% methanol.The sections were washed as in step 3 for 10 min in phosphate-buffered-saline (PBS, pH 7,3), and nonspecific protein-binding siteswere blocked with 3% normal goat serum (Camon, Wiesbaden,Germany) to reduce background staining. The sections were pro-cessed for immunlocalization of IGF-I, TGF-b2, bFGF (FGF-2), andEGF-R using a commercially available polyclonal antibody againstIGF-I (1:50 dilution; Labvision Corporation, Suffolk, UK), a poly-clonal antibody against TGF-b2 (1:50 dilution; Labvision Corpora-tion, Suffolk, UK), a polyclonal antibody against bFGF (1:200dilution; Santa Cruz Biotechnology, Santa Cruz, CA) and a poly-clonal antibody against EGF-R (1:200 dilution; Santa Cruz Biotech-nology, Santa Cruz, CA) as the primary antibodies. The optimumworking dilutions were determined by serial titration. After wash-ing with PBS, the specimens were incubated with biotinylated goatanti-polyvalent secondary antibody (Labvision Corporation, Suf-folk, UK) for IGF-1, TGF-b2 EGF-R and bFGF, for 30 min at roomtemperature. Streptomycin avidin-peroxidase conjugate (LabvisionCorporation, Suffolk, UK) was added and incubated for 30 min atroom temperature. The sections were visualized using 0.05%(w/v) 3,30-diaminobenzidine and 0.010% (v/v) hydrogen peroxidein PBS (10 mM, pH 7.4). These sections were counter-stained withhematoxylin and mounted with entellan. Control procedure wereundertaken to insure the specificity of immunoreaction. Negativecontrols were carried out by replacing the primary antibodies withPBS. Screen shots were taken with Olympus Digital camera (DP 20)attached at Olympus BX51 microscope. All the slides were exam-ined by the same observer who was blind to the tissue sections

K. Ergin et al. / Cytokine 43 (2008) 209–214 211

between the groups. The intensity of immunohistochemical stain-ing was graded semi-quantitatively as follows: (–) no immuno-staining; (+) weak staining; (++) moderate staining; (+++) strongstaining.

3. Results

The localization of IGF-I, TGF-b2, bFGF and EGF-R proteinsin cellular compartments in follicles is shown in Table 1and 2.

Granulosa cells of the primordial and primary follicles of new-born rat ovaries showed a moderate staining for IGF-I (Fig. 1A),while a strong staining for EGF-R in oocytes (Fig. 4A) was observed.However no staining for TGF-b2 and bFGF in oocytes and granulosacells (Figs. 2 and 3A) was observed.

When compared to newborn rats, a weak staining in granulosacells of primary follicles and no staining in granulosa cells ofsecondary follicles for IGF-1 was observed in one-month-old ratovaries. Primordial and primary follicles showed a strong stainingfor IGF-I in oocytes (data not shown) as well as secondary follicles

Fig. 1. IGF-I; (A) Newborn rat, moderate staining in granulosa cells of primary and primormonth-old rat, strong staining in the oocyte of a secondary follicle (asterisk), original m(asterisk) and a moderate staining in theca layer of a tertiary follicle (arrow), original m

Fig. 2. TGF-b2; (A) Newborn rat, no staining in follicles, original magnification: 400� (scafollicle (asterisk), original magnification: 400� (scale bars, 20 lm), (C) Adult rat, moderate(scale bars, 40 lm).

Fig. 3. bFGF; (A) Newborn rat, no staining in follicles, (B) One-month-old rat, strong stainthe oocyte of a secondary follicle (asterisk). Original magnification: 400� (scale bars, 20

(Fig. 1B). In addition a strong TGF-b2 and bFGF staining was seen inoocytes of the secondary follicles (Figs. 2 and 3B) and a moderateEGF-R staining in oocytes and granulosa cells of primary follicles(data not shown) and secondary follicles (Fig. 4B) was observedin one-month-old rats.

In comparison to one-month-old rats, oocytes of adult rat ova-ries showed a decreased staining for IGF-I (Fig. 1C), while a moder-ate IGF-I staining in theca cells of the secondary and tertiaryfollicles (Fig. 1C) was observed. Besides a moderate TGF-b2 stain-ing in oocytes of primary follicles (data not shown) and secondaryfollicles (Fig. 2C) was observed. Oocytes of the secondary follicles(Fig. 3C) and tertiary follicles (data not shown) showed a moderatestaining for bFGF in the adult group. Granulosa cells and oocytes ofadult rat ovaries showed a weak staining for EGF-R (Fig. 4C).

The stroma in one-month-old (Fig. 5A) and adult group (datanot shown) showed a strong staining for IGF-I, while a moderatestaining for TGF-b2 in one-month-old (data not shown) and adultgroup (Fig. 5B) was observed. A weak staining for bFGF in theone-month-old group (Fig. 5C), and a weak staining for EGF-R inthe adult group (Fig. 5D) was observed. The corpus luteii in the

dial follicles (arrowheads), original magnification: 400� (scale bars, 20 lm), (B) one-agnification: 400� (scale bars, 20 lm), (C) Adult rat, weak staining in the oocyte

agnification: 200� (scale bars, 40 lm).

le bars, 20 lm). (B) One-month-old rat, strong staining in the oocyte of a secondarystaining in the oocyte of a secondary follicle (asterisk), original magnification: 200�

ing in the oocyte of a secondary follicle (asterisk), (C) Adult rat, moderate staining inlm).

Fig. 4. EGF-R; (A) Newborn rat, strong staining in oocytes of primary follicles (arrowheads), original magnification: 400� (scale bars, 20 lm), (B) one-month-old rat, moderatestaining in the oocyte of a secondary follicle (asterisk), original magnification: 400� (scale bars, 20 lm), (C) Adult rat, weak staining in granulosa cells (arrowheads) and in theoocyte (asterisk) of a tertiary follicle. Original magnification: 200� (scale bars, 40 lm).

Fig. 5. Stroma cells; (A) Strong staining for IGF-I (asterisks) in one-month-old rat ovary, (B) Moderate staining for TGF-b2 (asterisks) in adult rat ovary, (C) Weak staining forbFGF (asterisk) in one-month-old rat ovary, (D) Weak staining for EGF-R (asterisks) in adult rat ovary. Original magnification: 100� (scale bars, 80 lm).

212 K. Ergin et al. / Cytokine 43 (2008) 209–214

adult group showed a moderate staining for IGF-I (Fig. 6A) and forTGF-b2 (Fig. 6B). However, no staining for bFGF and a weak stain-ing for EGF-R was observed (Fig. 6C and D).

4. Discussion

In this study, we intended to detect the expression and localiza-tion of IGF-I, TGF-b2, bFGF and EGF-R in rat ovaries of differentages by immunohistochemistry.

It is reviewed that IGF-I stimulated proliferation or differentia-tion of granulosa cells, depending on the stage of development ofthe follicle and it also stimulates proliferation of granulosa cellsfrom small follicles [16]. In addition it was claimed that IGF family

plays an important role in follicle development, dominant folliclegrowth, steroidogenesis and follicular atresia [17]. Previous studiesreported that IGF-I regulates human luteal steroidogenesis by anincrease in progesterone [18] and oestradiol production [19]. IGF-I was regarded as a co-gonadotrophin of LH in stimulating steroi-dogenesis in theca-intertitial cells [20]. In this study we found amoderate staining for IGF-I in granulosa cells of newborn rats, astrong staining in oocytes of young rats, a strong staining in stromacells of young and adult rats, a moderate staining in theca cells ofadult rats and also a moderate staining persisted in the corpusluteii. As it was seen, IGF-1 affected different parts of the ovary fol-licles in different ages. Hormonal changes with increasing age anddevelopment of the follicles might be the cause of these findings.

Fig. 6. Corpus luteii in the adult group (A) Moderate staining for IGF-I, (B) Moderate staining for TGF-b2, (C) No staining for bFGF, (D) Weak staining for EGF-R. Originalmagnification: 200� (scale bars, 40 lm).

K. Ergin et al. / Cytokine 43 (2008) 209–214 213

One family of peptides known to be involved in the regulationof ovarian development and function is the TGF-b superfamily[11]. Liu et al. were found that TGF-b increases the diameter of fol-licles from adult mice and has no effect on preantral follicles fromimmature mice [21]. TGF-b did not have an effect on preantral fol-licles from 11-day-old mice [21]. On the other hand, TGF-b stimu-lated the follicular growth of 56-day-old mice [21]. Interestingly,preantral follicles of 28-day-old mice showed an intermediate re-sponse, suggesting that the different responses are related to theprocess of physical maturity [21]. In comparison of newborn,young and adult rats, we observed TGF-b staining only in oocytesof young and adult rats in a variety of degrees. As a result of thesefindings, we thought that the effect of TGF-b does not appear untilpuberty.

It is reviewed by Knight et al. that members of the TGF-b super-family, are expressed in oocytes in a developmental-stage relatedmanner and act as intraovarian regulatory molecules [22]. Juengeland McNatty reported that TGF-b2 was found in the oocytes of sec-ondary follicles in rodents [23]. Similarly, we found stronger TGF-b2 staining in the oocytes of secondary follicles in young and adultrats. We suggested that TGF-b actions may be age and stage depen-dent. In the study of Chegini and Flanders, only small luteal cellsshowed intense immunostaining for TGF-b2; however, the large lu-teal cells had a weak immunostaining at midluteal phase in humanovary [24]. Besides, ovarian stromal cells do not immunostain forTGF-b2 [24]. In this study, we found a moderate staining of TGF-

b2 in the cells of the corpus luteii and of the stroma. The differenceof staining intensity might be due to different sizes of corpus luteiicells. Thus it could be the limitation of our study, not to focus onthe sizes of corpus luteii cells and stages of the luteal phase. Onthe other hand, compared to Chegini and Flanders’ study, the dif-ference of staining intensity in stromal cells might be because ofusing rat ovaries via human ovaries. In this respect, we could sug-gest that TGF-b2 may affect the oocyte development and corpusluteum survival.

The difference between the species might affect the expressionof growth factors such as bFGF. Ben-haroush et al. were detectedbFGF in granulosa cells from primary follicular stages onward ofadolescents/women, and in oocytes from fetuses and adults at allfollicular classes [12]. Besides Nilsson et al. were found that bFGFwas localized to primordial and early developing follicles and weremainly observed in the oocytes of 4-day-old rat ovaries which wereisolated and cultured for 14 days. In that study, elevated levels ofbFGF in the oocytes of primordial and early developing follicleswas described [13].

In our study, we found strong immunostaining in oocytes ofprimary and secondary follicles of young rats, moderate stainingin oocytes of secondary and tertiary follicles of adult rats and nostaining in newborn rats. Because we used different age groupsof rats such as newborn (0-day), young and adults, and studiedonly immunohistochemically, our findings were not in accor-dance with Nilsson’s study. In addition to this, our findings

214 K. Ergin et al. / Cytokine 43 (2008) 209–214

showed that bFGF affected follicles especially in young and adultrats.

In some studies EGF-receptor (EGF and TGFa bind to this recep-tor) has been identified in granulosa cells, corpus luteum and oo-cytes [25,26]. Dark staining was found in the oocytes comparedto follicular and stromal cells in human fetuses [25]. Bennet et al.postulated that EGF-R may act as local regulators for the develop-ment or degeneration of individual oocytes [25]. We found a strongstaining for EGF-R in oocytes of newborn rats, a moderate stainingin one-month-old rats and a weak staining in adult rats. From thispoint of view, a decreased staining of EGF-R was seen by increasingage. Therefore, we thought that the effect of EGF-R were decreasingwith age and hence EGF-R might have the strongest effects in new-born term.

Ayyagari and Khan-Dawood demonstrated the presence of EGF-receptors in human corpus luteum [26]. We found a weak stainingin corpus luteii for EGF-R. Since EGF-R staining decreased with agein oocytes, this was an expected result.

In conclusion, we have demonstrated the presence of IGF-I, TGF-b2, bFGF and EGF-R proteins in developing ovaries, suggesting thatgrowth factors play a pivotal role in ovarian function. The role ofgrowth factors in controlling degeneration or growth (or both) ofovary follicles remain as explained. Therefore, additional studiesshould be undertaken to elucidate the role of these and other pro-teins in development and atresia of ovarian follicles.

Acknowledgments

The authors thank the Adnan Menderes University ResearchFund for financial support (TPF-4015) for this study and Assist.Prof. Dr. Yüksel Yıldız for his language assistance.

References

[1] Hsueh AJ, Billig H, Tsafriri A. Ovarian follicle atresia: a hormonally controlledapoptotic process. Endocr Rev 1994;15:707–24.

[2] Quirk SM, Cowan RG, Harman RM, Hu CL, Porter DA. Ovarian follicular growthand atresia: the relationship between cell proliferation and survival. J Anim Sci2004;82(E-Suppl):E40–52.

[3] Kaiser GG, Sinowatz F, Palma GA. Effects of growth hormone on femalereproductive organs. Anat Histol Embryol 2001;30:265–71.

[4] Ulug U, Turan E, Tosun SB, Erden HF, Bahceci M. Comparison of preovulatoryfollicular concentrations of epidermal growth factor, insulin-like growthfactor-I, and inhibins A and B in women undergoing assisted conceptiontreatment with gonadotropin-releasing hormone (GnRH) agonists and GnRHantagonists. Fertil Steril 2007;87:995–8.

[5] Luciano AM, Modina S, Gandolfi F, Lauria A, Armstrong DT. Effect of cell-to-cellcontact on in vitro deoxyribonucleic acid synthesis and apoptosis responses ofbovine granulosa cells to insulin-like growth factor-I and epidermal growthfactor. Biol Reprod 2000;63:1580–5.

[6] Monget P, Fabre S, Mulsant P, Lecerf F, Elsen JM, Mazerbourg S, et al. Regulationof ovarian folliculogenesis by IGF and BMP system in domestic animals.Domest Anim Endocrinol 2002;23:139–54.

[7] Mazerbourg S, Bondy CA, Zhou J, Monget P. The insulin-like growth factorsystem: a key determinant role in the growth and selection of ovarianfollicles? a comparative species study. Reprod Domest Anim2003;38:247–58.

[8] Villavicencio A, Iñiguez G, Johnson MC, Gabler F, Palomino A, Vega M.Regulation of steroid synthesis and apoptosis by insulin-like growth factor Iand insulin-like growth factor binding protein 3 in human corpus luteumduring the midluteal phase. Reproduction 2002;124(4):501–8.

[9] Nikolettos N, Asimakopoulos B, Köster F, Schöpper B, Scultz C, Caglar GS, et al.Cytokine profile in cases with premature elevation of progesterone serumconcentrations during ovarian stimulation. Physiol Res 2007;8. [Epub ahead ofprint].

[10] Monniaux D, Monget P, Besnard N, Huet C, Pisselet C. Growth factors andantral follicular development in domestic ruminants. Theriogenology1996;47:3–12.

[11] Drummond AE. TGF beta signalling in the development of ovarian function.Cell Tissue Res 2005;322(1):107–15.

[12] Ben-Haroush A, Abir R, Ao A, Jin S, Kessler-Icekson G, Feldberg D, et al.Expression of basic fibroblast growth factor and its receptors in human ovarianfollicles from adults and fetuses. Fertil Steril 2005;849(Suppl 2):1257–68.

[13] Nilsson E, Parrott JA, Skinner MK. Basic fibroblast growth factor inducesprimordial follicle development and initiates folliculogenesis. Mol CellEndocrinol 2001;175(1–2):123–30.

[14] Chen YG, Meng AM. Negative regulation of TGF-beta signaling in development.Cell Res 2004;14(6):441–9.

[15] Méndez C, Alcántara L, Escalona R, López-Casillas F, Pedernera E. Transforminggrowth factor beta inhibits proliferation of somatic cells without influencinggerm cell number in the chicken embryonic ovary. Cell Tissue Res2006;325(1):143–9.

[16] Monget P, Fabre S, Mulsant P, Lecerf F, Elsen JM, Mazerbourg S, et al.Regulation of ovarian folliculogenesis by IGF and BMP system in domesticanimals. Domest Anim Endocrinol 2002;23(1–2):139–54.

[17] Giudice LC. Insulin-like growth factor family in Graafian follicle developmentand function. J Soc Gynecol Investig 2001;8:S26–9.

[18] Devoto L, Kohen P, Castro O, Vega M, Troncoso JL, Charreau E. Multihormonalregulation of progesterone synthesis in cultured human midluteal cells. J ClinEndocrinol Metab 1995;80(5):1566–70.

[19] Johnson MC, Devoto L, Retamales I, Kohen P, Troncoso JL, Aguilera G.Localization of insulin-like growth factor (IGF-I) and IGF-I receptorexpression in human corpora lutea: role on estradiol secretion. Fertil Steril1996;65(3):489–94.

[20] Duleba AJ, Spaczynski RZ, Olive DL, Behrman HR. Divergent mechanismsregulate proliferation/survival and steroidogenesis of theca-interstitial cells.Mol Hum Reprod 1999;5(3):193–8.

[21] Liu X, Andoh K, Abe Y, Kobayashi J, Yamada K, Mizunuma H, et al. Acomparative study on transforming growth factor-beta and activin A forpreantral follicles from adult, immature, and diethylstilbestrol-primedimmature mice. Endocrinology 1999;140(6):2480–5.

[22] Knight PG, Glister C. Local roles of TGF-beta superfamily members in thecontrol of ovarian follicle development. Anim Reprod Sci 2003;78(3–4):165–183.

[23] Juengel JL, McNatty KP. The role of proteins of the transforming growth factor-beta superfamily in the intraovarian regulation of follicular development. HumReprod Update 2005;11(2):143–60. [Epub 2005 Feb 10].

[24] Chegini N, Flanders KC. Presence of transforming growth factor-beta and theirselective cellular localization in human ovarian tissue of various reproductivestages. Endocrinology 1992;130(3):1707–15.

[25] Bennett RA, Osathanondh R, Yeh J. Immunohistochemical localization oftransforming growth factor-alpha, epidermal growth factor (EGF), and EGFreceptor in the human fetal ovary. J Clin Endocrinol Metab 1996;81(8):3073–6.

[26] Ayyagari RR, Khan-Dawood FS. Human corpus luteum: presence of epidermalgrowth factor receptors and binding characteristics. Am J Obstet Gynecol1987;156(4):942–6.