Cell Tissue Res (1992) 270:87-93
Cell&Tissue Research 9 Springer-Verlag 1992
Ovarian innervation develops before initiation of follieulogenesis in the rat Sasha Malamed 1, Jean A. Gibney 1, and Sergio R. Ojeda 2 1 Department of Neuroscienceand Cell Biology,UMDNJ - Robert Wood Johnson Medical School, Piscataway, New Jersey 08854, USA 2 Division of Neuroscience, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue Beaverton, OR 97006, USA Received January 3, 1992 / Accepted May 7, 1992
Summary. Sympathetic neurotransmitters have been shown to be present in the ovary of the rat during early postnatal development and to affect steroidogenesis before the ovary becomes responsive to gonadotropins, and before the first primordial follicles are formed. This study was undertaken to determine if development of the ovarian innervation is an event that antedates the initiation of folliculogenesis in the rat, Rattus norvegicus. Serial sections of postnatal ovaries revealed a negligible frequency of follicles 24 h after birth (about 1 primordial follicle per ovary). Twelve hours later there were about 500 follicles per ovary, a number that more than doubled to about 1300 during the subsequent 12 h, indicating that an explosive period of follicular differentiation occurs between the end of postnatal days 1 and 2. Electron microscopy demonstrated that before birth the ovaries are already innervated by fibers containing clear and dense-core vesicles. Immunohistochemistry performed on either fetal (day 19) or newborn (less than 15h after birth) ovaries showed the presence of catecholaminergic nerves, identified by their content of immunoreactive tyrosine hydroxylase (TH), the rate-limiting enzyme in catecholamine biosynthesis. While some of these fibers innervate blood vessels, others are associated with primordial ovarian cells, thereby suggesting their participation in non-vascular functions. Since prefollicular ovaries are insensitive to gonadotropins, the results suggest that the developing ovary becomes subjected to direct neurogenic influences before it acquires responsiveness to gonadotropins. Key words: Ovarian nerves - Development - Folliculogenesis - Tyrosine hydroxylase - Immunohistochemistry - Electron microscopy - Rat (Sprague Dawley)
Numerous reports have appeared describing the intragonadal distribution of ovarian nerves in mammals (for Correspondence to: S. Malamed
reviews see Burden 1978; Baljet and Drucker 1979; Mohsin and Pennefather 1979) and identifying the putative neurotransmitters presumably employed by the innervating fibers to regulate ovarian function (reviewed by Burden 1985; Ojeda etal. 1989; Ojeda and Lara 1989). In spite of this, the developmental phase during which the ovaries become innervated and, more importantly, the temporal and functional relationship that this event may have with the initiation of folliculogenesis (Hirshfield J 991) have not been defined. Resolution of this issue is important because it may provide new insights into potential neuroendorine mechanisms that may contribute to the initiation of follicle formation. Substantial evidence exists that early follicular development is a pituitary-independent phenomenon (Hertz 1963; Peters et al. 1973; Schwartz 1974; Funkenstein et al. 1980), and that responsiveness to gonadotropins is not acquired until the end of the first week of postnatal life (Ben-Or 1963 ; Lamprecht et al. 1973 ; Funkenstein et al. 1980; Sokka and Huhtaniemi 1990). In contrast, the ovary responds to activation of adenylate cyclase (Lamprecht et al. 1973; George and Ojeda 1987; Sokka and Huhtaniemi 1990) and to cyclic AMP itself (George and Ojeda 1987) already during fetal life (George and Ojeda 1987; Sokka and Huhtaniemi 1990). Other studies have demonstrated the ability of vasoactire intestinal peptide (VIP), a putative neurotransmitter/neuromodulator contained in ovarian nerves (Papka etal. 1985; Ahmed etal. 1986; Kannisto etal. 1986), to induce cAMP formation and aromatase activity in fetal ovaries (George and Ojeda 1987). Since gonadotropin receptors are undetectable before postnatal day 4 (Peluso et al. 1976; Siebers et al. 1977; Smith White and Ojeda 1981; Sokka and Huhtaniemi 1990), these observations suggest that the first regulatory influences received by the gland during early development may be neurogenic rather than hormonal. To begin examining this hypothesis it is first required to determine if prefollicular ovaries are innervated. The present study addresses this issue. A preliminary report of these findings has appeared (Malamed et al. 1990).
88
1600 1308:1:217 >, 1 4 0 0 0
1200 r
_r 0
~_ 1 0 0 0
2 800
E 5 0 2 +- 9 9 '-
600
E 400
.~_ LU 2 0 0 1.25+0.75 1.0
1.5 Days after birth
Fig. 1 a-d. Sections of neonatal rat ovaries showing the sharp increase in n u m b e r of primordial follicles that occurs between 24 to 48 h after birth, a Histological aspect of primordial follicles characterized by single layer of flattened somatic cells surrounding each oocyte, • b 24 h ovary, x 4 2 0 ; e 36 h ovary, x 4 2 0 ; d 48 h ovary, x 420. Arrows indicate primordial follicles Fig. 2. Development of primordial follicles in neonatal rat ovary. Columns represent mean values. Vertical lines indicate standard errors of the means, n = 4 for each column Fig. 3a, b, Immunocytochemical localization of tyrosine hydroxylase (TH). a T H immunoreactivity in branching nerve fiber in section of 19-day-old fetal rat ovary. Hilum of ovary in upper left corner; border of ovary in lower right corner. Arrows indicate cell clusters in medullary region, x 210. b TH immunoreactivity in branching nerve fiber in section of rat ovary less than 15 h after birth, x 390
20
89
Fig. 4. Transverse section of nerve fiber (N) encapsulated by ovarian cell in 22-day-old fetal ovary. Open arrows Round and elongate clear vesicles; small arrowhead clear vesicle whose membrane is continuous with plasma membrane of nerve fiber; long arrow paired plasma membrane thickenings without vesicles nearby; large arrowhead region of invagination of ovarian cell plasma membrane which envelops nerve fiber. M Mitochondrion of nerve fiber; m mitochondrion of ovarian cell; O ovarian cell. x 50000. Bar length equals 1 gm Fig. 5. Longitudinal section of two nerve fibers in 22-day-old fetal ovary. Open arrows Round and elongate clear vesicles; short arrows densecore vesicles; long arrows paired plasma membrane thickenings. These membrane appositions couple nerve fiber (N) to nerve fiber and nerve fiber to endothelial cell (E). Thin bars adjacent to bundles of neurofilaments. O Ovarian cell. x 31400. Thick bar: 1 g M
Materials and methods Fetal ovaries were obtained from litters of timed-pregnant Sprague Dawley rats, Rattus norvegicus (Bantin and Kingman, Fremont, Calif.). Postnatal ovaries were from Long-Evans derived rats (bred in Dr. Malamed's laboratory). Sprague Dawley and Long-Evans rats have the same gestation period (20-22 days). The exact postnatal ages of the pups were accurately determined by carefully monitoring the times of birth. The number of follicles per ovary was estimated as follows: The ovaries were first fixed in Gomori 1-2-3 fixative (Humason 1972) at room temperature. Thereafter, they were dehydrated with
ethanol, embedded in paraffin, serially sectioned at 5 gm and stained with hematoxylin and eosin. A primordial follicle was defined as an oocyte surrounded by a single incomplete or complete layer of flattened somatic cells (Fig. 1A). A primary follicle was one with cuboidal follicle cells; few of these appeared and only in 48 h-old ovaries. For 24 (_+ 1) h-old ovaries, counts were made of the numbers of follicles in every section of each ovary. For each 36 ( + 1 ) and 48 (+1) h-old ovary, counts were made of the total numbers of follicles present in every 5th or 6th section of the total number of sections per ovary; the mean total number of follicles per section was then multiplied by the total number of sections of the ovary to obtain the total number of follicles
90 per ovary. Values obtained by this method tend to be spuriously high because the thickness of each section is about one third the mean diameter of a primordial follicle as determined with an objective micrometer and eyepiece graticule. Thus, each primordial follicle appears on more than one section. However, with this method follicles with oocytes that are incompletely surrounded by follicle cells are not counted in sections in which the follicle cells do not appear. This tends to correct the overestimation. Regardless of the validity of the estimated absolute values for numbers of follicles per ovary, the uniform application of the method of calculation to the data from ovaries of all ages examined should produce valid relative values. For electron microscopy, the ovaries were immersed in 50% Karnovsky's fixative (Karnovsky 1965) for 4 h at 4~ C. Post-fixation was in 1% OsO4 in 0.1 M cacodylate buffer, pH 7.4 for 1 h at 0~ C. Uranyl acetate (1% aqueous) for 1 h at room temperature was used for en bloc staining followed by routine ethanol dehydration and embedding in EMbed 812 (Electron Microscopy Sciences, Fort Washington, Pa.). Tissue blocks were sectioned at 70-80 nm thickness and examined in a Philips CM12 electron microscope. For immunohistochemical studies, neonatal rats were anesthetized with ether and perfnsed with 20-25 ml of 0,9% saline solution containing heparin (20 units/ml), 0.008% Lid-o-cain (The Butler Company, Columbus, Ohio) and 0.02% NaNO2 for 3 min. Thereafter, the perfusion was continued at the same flov~ rate with a mixture of 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for 7 rain. The ovaries were excised and immersed in the same fixative for 8 min. They were transferred to 4% paraformaldehyde in 0.1 M phosphate buffer, pH 9.0, for 4 h at 4~ and left overnight in phosphate buffer, pH 7.4, containing 20% sucrose at 4~ after which 20 gm cryostat sections were cut and mounted on poly-l-lysine (Sigma Chemical Company, St. Louis, Mo.) coated slides. Rabbit anti-bovine adrenal tyrosine hydroxylase (TH) (Eugene Tech International, Inc, Allendale, N.J.) at 1 : 500-1 : 5000 dilutions was used for all immunocytochemical studies. For negative controls the primary antiserum was replaced by normal goat serum. Adrenal glands from the same rats that provided the ovaries were used as positive controls. Incubation was overnight at room temperature or, with lower concentrations of antiserum, for 67 h (over the weekend). The ovary sections were prepared according to the Avidin DH: biotinylated horseradish peroxidase H complex (ABC) method (Vector Laboratories, Burlingame, Calif.).
Results At one day after birth ovarian follicles were essentially absent (Figs. 1, 2). In the ensuing 24 h, follicle formation increased sharply to about 500 follicles per ovary at 36 h after birth, and to about 1300 follicles per ovary 12 h later. Fig. I b shows the absence o f follicles 24 h after birth, and the marked increase in follicle number during the next 24 h (Fig. i c, d). Fig. 2 provides the quantitative data for these changes. The follicles present were primordial follicles except for occasional primary follicles ( < 5%) in the 48 h ovaries. The latter had cuboidal follicle cells rather than the flattened follicle cells of the primordial follicle. In contrast to the absence of follicles in newborn ovaries, nerve fibers were detected even during fetal life. Immunocytochemical detection of T H at the light-microscopy level provided evidence for the adrenergic nature of these nerve fibers in the ovary from either fetal (19-day-old, Fig. 3a) or newborn ( < 1 5 h - o l d ) rats (Fig. 3 b) prior to the time when primordial follicles ap-
pear. The TH-labeled fibers observed were in the hilum or medullary part of the ovary; none were found in the cortex. Figs. 4 and 5 show the ultrastructural appearance of nerve fibers identified in a 22-day-old fetal ovary. A cross-section of a fiber suggests that it is enveloped by an ovarian cell (Fig. 4). These early nerve fibers are endowed with dense core and clear vesicles (Figs. 4, 5). Two size populations of pleomorphic clear vesicles were seen. Clear vesicles with a mean diameter of 51.3___ 5.8 nm appear in Fig. 4 and with a mean diameter of 79.3 _+8.2 nm in Fig. 5 (P < 0.01 ; Student's t-test). Some o f these vesicles are elongate; their nominal diameters were calculated by using the means of their long and short dimensions. Dense-core vesicles, most of which are elongate, have a mean diameter of 147 + 14 nm. The mean dimensions of the elongate dense core vesicles are 185_+23 n m x 122_+14 nm. These appear in the same fibers that contain the larger clear vesicles (Fig. 5), but are not co-localized with the smaller clear vesicles (Fig. 4). Paired plasma membrane thickenings were observed; these involve nerve to ovarian cell (Fig. 4), nerve to nerve, and nerve to endothelial cell appositions (Fig. 5). The prenatal nerve fibers appear in the formative ovarian tissue proper (Fig. 4) or are associated with blood vessels (Fig. 5). The frequency of TH-labeled fibers is low; thorough examination by light microscopy of all sections of an ovary revealed up to 6 fibers per ovary. In contrast, T H labeling is abundant in adrenal glands from the same animals that provided the ovaries (not shown). T H labeling was absent in ovary and adrenal gland sections incubated without antibody to TH.
Discussion Although substantial progress has been made towards the understanding of mature ovarian function, little if anything - is known about the regulatory signals that govern early ovarian development. Ovarian function does not become subjected to gonadotropin control until approximately the end of the first postnatal week of life (Lamprecht etal. 1973; Sokka and Huhtaniemi 1990). This refractoriness appears to result mostly from a lack of gonadotropin receptors (Peluso etal. 1976; Siebers et al. 1977; Smith White and Ojeda 1981; Sokka and Huhtaniemi 1990). In other studies (Picon et al. 1985; George and Ojeda 1987), the c A M P analog dibutyryl c A M P was found to increase ovarian aromatase activity even during fetal life, suggesting that the intracellular transduction mechanism that mediates gonadotropin actions in the ovary is intact long before the gonad becomes responsive to either L H or F S H (Picon et al. 1985; George and Ojeda 1987; Sokka and Huhtaniemi 1990). Gonadotropins play a major role in the control of ovarian development and secretory activity, but they are not the only substances that affect ovarian function through activation of cAMP-dependent mechanisms. Norepinephrine and VIP, two neurotransmitters con-
91 tained in ovarian nerves (Burden 1978; Mohsin and Pennefather 1979; Papka et al. 1985; Ahmed et al. 1986; Kannisto et al. 1986; Klein and Burden 1988) stimulate steroidogenesis via the cAMP-mediated transduction pathway (Davoren and Hsueh 1984; Zor and Kliachko 1985). Both have been detected in the neonatal ovary between the second and fifth postnatal day (Ben-Jonathan et al. 1984; Advis et al. 1989), i.e., before the gland acquires responsiveness to gonadotropins. Moreover, VIP was found to stimulate cAMP formation and induce aromatase activity in fetal ovaries, which are unresponsive to LI-I and/or FSH (George and Ojeda 1987). These findings indicate that the immature rat ovary becomes exposed to putative neurotransmitters and develops the capacity to respond to neurotransmitter stimulation during an early, gonadotropin-independent phase of development. Although the ovary may have the intrinsic ability to synthesize catecholamines (Dissen et al. 1990) and VIP (Gozes and Tsafriri 1986), a substantial fraction of these transmitters reaches the ovary via the extrinsic innervation (Lawrence and Burden 1980; Dees et al. 1986). The present study provides ultrastructural evidence that the innervation of the ovary is a fetal event that antedates the postnatal initiation of folliculogenesis. It also demonstrates, by means of immunohistochemistry, that some of the prenatal innervating fibers are catecholaminergic as determined by the presence of immunoreactive TH, the rate-limiting enzyme in catecholamine biosynthesis. The TH-positive fibers are sparse. Thorough light-microscopic examination of all the 20-30 serial sections of each of the 16 ovaries studied immunocytochemically revealed, at most, only six TH-labeled fibers per ovary. However, the same fiber may have appeared in more than one section, and the plane of section may have revealed separate segments of the same nerve fiber. Thus, it is likely that the actual number of nerve fibers per ovary is fewer than six. The existence of ovarian intercellular gap junctions (Burghardt and Anderson 1981) suggests a mechanism by which these initial neural inputs may affect early ovarian function. The electron-microscopic studies revealed the presence of synaptoid appositions of plasma membrane thickenings in the ovarian nerves. There are, however, no vesicles associated with the thickenings suggesting that these paired regions of plasma membranes do not serve for the transmission of nerve impulses (Peters et al. 1991). Also, because the nerve fiber pairings involve an endothelial cell and an ovarian cell as well as another nerve fiber, it is possible that these structures are merely areas of adhesion (puncta adhaerentia). The TH-labeled fibers may be assumed to be efferents. If, as is likely, these autonomic fibers are postganglionic, highly specialized synapses would not be expected. Ovarian nerve terminals resemble other autonomic nerve terminals which are not typically in intimate contact with the cells they influence (Burden 1978; Smolen 1988; Weiss 1988; Nilsson 1983). These terminals (as shown in Fig. 5) may lie within a trough or deep groove formed by the effector cell membrane (Dahl 1970; Fink and Schofield 1971; Nilsson 1983; Malamed et al. 1984). This arrangement
may be an adaptation for targeting the delivery of neurotransmitters to the effector cell. Among the ultrastructural features apparent in the nerve fibers of the fetal ovary are several types of vesicles. Two types of nerve fibers were observed; one type with small, clear, round and elongate vesicles, the other with larger, clear, round and elongate vesicles and large dense-core vesicles. The smaller clear vesicles with a mean diameter of 51 nm were seen in a fiber without dense-core vesicles. The morphological features of the smaller clear vesicles suggest that the fiber containing them is inhibitory, possibly GABAergic (Nilsson 1983; Weiss 1988; Peters et al. 1991). The larger clear vesicles with a mean diameter of 79 nm were in a fiber that also contained round and elongate profiles of dense-core vesicles. The dense-core vesicles (147 nm) observed are most likely peptidergic (Nilsson 1983 ; Weiss 1988 ; Peters et al. 1991). The abrupt initiation of folliculogenesis and the brevity of the interval required for the formation of hundreds of new follicles is of considerable interest and in agreement with recent observations by other authors (Rajah and Hirshfield 1991). The presence of neurotransmitters in pre-follicular ovaries suggested by the present study raises the possibility that the initial formation of follicles is, at least in part, facilitated by signals of neural origin. In addition to direct neural influences, factors related to the development of innervation may contribute to the process of folliculogenesis. For instance, blockade of nerve growth factor (NGF) action at the time of birth has been shown to delay follicular development (Lara et al. 1990). Other reports have described the presence of N G F receptors in thecal cells of ovarian follicles (Dissen et al. 1991) and neurotrophin mRNA expression in granulosa cells and oocytes (Ernfors et al. 1990; Hallbook et al. 1991). These observations raise the intriguing possibility that initiation of folliculogenesis involves the participation of molecules until now thought to be exclusively neurotrophic. Such a role is suggested by recent experiments in which we have found that in vivo or in vitro exposure of neonatal ovaries to antibodies to NGF results in a dramatic reduction in follicular formation (Dissen et al. 1992). Indeed, neurotrophins may be required for the differentiation of other non-neural developing tissues such as kidney tubules (Sariola et al. 1991). The present results do not provide direct evidence in support of either a neural or atrophic factor-dependent mechanism influencing the initiation of folliculogenesis, but conclusively demonstrate that development of innervation precedes both the formation of follicles and the acquisition of responsiveness to gonadotropins at the onset of ovarian maturation.
Acknowledgements. We thank Ms. Janie Gliessman for editorial assistance, and Dr. GeoffreyMcAuliffe,Dr. David Crockett, Mr. Alvin Sharma, Mr. Nadeem Haider, and Ms. Maria E. Costa for technical assistance. This work was supported by NIH Grants HD24870, RR-00163 and HD-18185. This is publication no. 1841 of the Oregon RegionalPrimate Research Center.
92 References Advis JP, Ahmed CE, Ojeda SR (1989) Direct hypothalamic control of vasoactive intestinal peptide (VIP) levets in the developing rat ovary. Brain Res Bull 22:605-610 Ahmed CE, Dees WL, Ojeda SR (1986) The immature rat ovary is innervated by vasoactive intestinal peptide-containing fibers and responds to VIP with steroid secretion. Endocrinology 118:1682-1689 Baljet B, Drukker J (1979) The extrinsic innervation of the abdominal organs in the female rat. Acta Anat 104:243-267 Ben-Jonathan N, Arbogast LA, Rhoades TA, Bahr J (1984) Norepinephrine in the rat ovary: ontogeny and de novo synthesis. Endocrinology 115 : 1426-1431 Ben-Or S (1963) Morphological and functional development of the ovary of the mouse. I. Morphology and histochemistry of the developing ovary in normal conditions and after FSH treatment. J Embryol Exp Morph 11 : 1-11 Burden HW (1978) Ovarian innervation. In: Jones RE (ed) The vertebrate ovary: comparative biology and evolution. Plenum Press, New York, pp 615-638 Burden HW (1985) The adrenergic innervation of mammalian ovaries. In: Ben-Jonathan N, Bahr JM, Weiner RI (eds) Catecholamines as hormone regulators. Raven Press, New York, pp 261-278 Burghardt RC, Anderson E (1981) Hormonal modulation of gap junctions in rat ovarian follicles. Cell Tissue Res 214:181-193 Dahl E (1970) Studies of the fine structure of ovarian interstitial tissue. 3. The innervation of the thecal gland of the domestic fowl. Z Zellforsch 109:212-226 Davoren BJ, Hsueh AJW (1984) Vasoactive intestinal peptide: a novel stimulator of steroidogenesis by cultured rat granulosa cells. Biol Reprod 33 : 37-52 Dees WL, Ahmed CE, Ojeda SR (1986) Substance P (SP) and vasoactive intestinal peptide (VIP)-containingnerve fibers reach the ovary by independent routes. Endocrinology 119:638-641 Dissen GA, Nilaver G, Hill DF, Ojeda SR (1990) Expression of nerve growth factor receptor (NGFrec) and tyrosine hydroxylase (TH) mRNAs in the immature rat ovary (abstract). Soc Neurosci 16:296 Dissen GA, Hill DF, Costa ME, Ma YJ, Ojeda SR (1991) Nerve growth factor receptors in the prepubertal rat ovary. Mol Endocrinol 5 : 1642-1650 Dissen GA, Malamed S, Gibney JA, Hirshfield AN, Ojeda SR (1992) Neurotrophins are required for follicular formation in the mammalian ovary. Soc Neurosci Abstr 18 (in press) Ernfors P, Wetmor C, Olson L, Persson H (1990) Identification of cells in rat brain and peripheral tissues expressing mRNA for members of the nerve growth factor family. Neuron 5:511526 Fink G, Schofield GC (1971) Experimental studies on the innervation of the ovary in cats. J Anat 109 : 115-126 Funkenstein B, Nimrod A, Lindner HR (1980) The development of steroidogenie capability and responsiveness to gonadotropins in cultured neonatal rat ovaries. Endocrinology 106: 98-106 George FW, Ojeda SR (1987) Vasoactive intestinal peptide induces aromatase activity before development of primary follicles and responsiveness to FSH. Proc Natl Acad Sci USA 84:5803-5807 Gozes I, Tsafriri A (1986) Detection of vasoactive intestinal peptide-encoding messenger ribonucleie acid in the rat ovaries. Endocrinology 119: 2606-2610 Hallbook F, Ibanez CF, Persson H (1991) Evolutionary studies of the nerve growth factor family reveal a novel member abundantly expressed in Xenopus ovary. Neuron 6 : 845-858 Hertz R (1963) Pituitary independence of the prepubertal development of the ovary of the rat and the rabbit and its pertinence to hypo-ovarianism in women. In: Grady HG, Smith DE (eds) The ovary. Williams & Wilkins, Baltimore, pp 120 127 Hirshfield AN (1991) Development of follicles in the mammalian ovary. Int Rev Cytol 123:43-101
Humason GL (1972) Animal Tissue Techniques. Freeman, San Francisco, p 15 Kannisto P, Ekblad E, Helm G, Owman C, Sjoberg NO, Stjernquist M, Sundler F, Walles B (1986) Existence and co-existence of peptides in nerves of the mammalian ovary and oviduct demonstrated by immunocytochemistry. Histochemistry 86:25 34 Karnovsky MJ (1965) A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 27:137 Klein CM, Burden HW (1988) Substance P- and vasoactive intestinal polypeptide (VIP)-immunoreactive nerve fibers in relation to ovarian post-ganglionic perikarya in para- and pre-vertebral ganglia: evidence from combined retrograde tracing and immunocytochemistry. Cell Tissue Res 252:403-410 Lamprecht SA, Zor U, Tsafriri A, Lindner HR (1973) Action of prostaglandin E2 and of luteinizing hormone on ovarian adenylate cyclase, protein kinase and ornithine decarboxylase activity during postnatal development and maturity in the rat. J Endocrinol 57:217-233 Lara HE, McDonald JK, Ojeda SR (1990) Involvement of nerve growth factor in female sexual development. Endocrinology 126:364-375 Lawrence IE Jr, Burden HW (1980) The origin of the extrinsic adrenergic innervation of the ovary. Anat Rec 196:51-59 Malamed S, Gibney JA, Yamasaki DS, Carsia RV (1984) Ultrastructural evidence for adrenocortical-neuronal junctions (abstract). Anat Record 208 : 113A Malamed S, Gibney JA, Ojeda SR (1990) Ovarian innervation develops before initiation of folliculogenesis. Anat Rec 226 : 64A65A Mohsin S, Pennefather JN (1979) The sympathetic innervation of the mammalian ovary. A review of pharmacological and histological studies. Clin Exp Pharmacol Physiol 6:335-354 Nilsson (1983) Autonomic nerve function in the vertebrates. Springer, Berlin, pp 43-44 Ojeda SR, Lara H (1989) The role of the sympathetic nervous system in the regulation of ovarian function. In: Pirke KM, Wuttke W, Schweiger U (eds) The menstrual cycle and its disorders. Springer, Berlin, pp 26-32 Ojeda SR, Lara H, Ahmed CE (1989) The potential relevance of vasoactive intestinal peptide to ovarian physiology. In: Adashi E (ed) Putative intraovarian regulators. Seminars in reproductive endocrinology, vol 7. Thieme, New York, pp 52-60 Papka RE, Cotton JP, Traurig HA (1985) Comparative distribution of neuropeptide tyrosine-, vasoactive intestinal peptide-, substance P-immunoreactive, acetylcholinesterase-positive and noradrenergic nerves in the reproductive tract of the female rat. Cell Tissue Res 242:475-490 Peluso JJ, Steger RW, Hafez ESE (1976) Development of gonadotropin-binding sites in the immature rat ovary. J Reprod Fertil 47 : 55-58 Peters H, Byskov AG, Eaher M (1973) Intraovarian regulation of follicle growth in the immature mouse. In: Peters H (ed) The development and maturation of the ovary and its functions. Excerpta Medica, Amsterdam, pp 20-23 Peters A, Palay SL, Webster HdeF (1991) The Fine Structure of the Nervous System, 3 edn. Oxford University Press, New York, pp 8, 169-186 Picon R, Pelloux MC, Benhaim A, Gloaguen F (1985) Conversion of androgen to estrogen by the rat fetal and neonatal female gonad: effects of dcAMP and FSH. J Steroid Biochem 23 : 9951000 Rajah R, Hirshfield AN (1991) The changing architecture of the rat ovary during the immediate postpartum period: a three dimensional (3D) reconstruction. Biol Reprod 44 [Suppl 1] : 152 Sariola H, Saarma M, Sainio K, Arumae U, Palgi J, Vaahtokari A, Thesleff I, Karavanov A (1991) Dependence of kidney morphogenesis on the expression of nerve growth factor receptor. Science 254:571-573
93 Schwartz NB (1974) The role of FSH and LH and of their antibodies on follicle growth and on ovulation. Biol Reprod 10 : 236-272 Siebers JW, Schmidkke J, Engel W (1977) hCG-insensitivity of the postnatal ovary is due to the lack of a specific receptor. Experientia 33 : 689 Smith White S, Ojeda SR (1981) Changes in ovarian luteinizing hormone (LH) and follicle stimulating hormone (FSH) receptor content and in gonadotropin-induced ornithine decarboxylase activity during prepubertal and pubertal development of the rat. Endocrinology 109:152-161
Smolen AJ (1988) Morp])ology of synapses in the autonomic nervous system. J E1 Micr Tech 10:187 204 Sokk a T, Huhtaniemi I (1990) Ontogeny of gonadotropin receptors and gonadotropin-stimulated cyclic AMP production in the neonatal ovary. J Endocrinol 127:297-330 Weiss L (1988) Cell and tissue, 6th edn. Urban and Schwarzenberg, Baltimore, pp 300-302 Zor U, Kliachko S (1985) The fl-adrenergic system in rat ovarian granulosa cells: hormonal and prostaglandin dependence in viyo and spontaneous development in vitro. In: Ben-Jonathan N I Bahr JM, Weiner RI (eds) Catecholamines as hormone regulators. Raven Press, New York, pp 311-328
Cell&Tissue Research 9 Springer-Verlag 1992
Ovarian innervation develops before initiation of follieulogenesis in the rat Sasha Malamed 1, Jean A. Gibney 1, and Sergio R. Ojeda 2 1 Department of Neuroscienceand Cell Biology,UMDNJ - Robert Wood Johnson Medical School, Piscataway, New Jersey 08854, USA 2 Division of Neuroscience, Oregon Regional Primate Research Center, 505 N.W. 185th Avenue Beaverton, OR 97006, USA Received January 3, 1992 / Accepted May 7, 1992
Summary. Sympathetic neurotransmitters have been shown to be present in the ovary of the rat during early postnatal development and to affect steroidogenesis before the ovary becomes responsive to gonadotropins, and before the first primordial follicles are formed. This study was undertaken to determine if development of the ovarian innervation is an event that antedates the initiation of folliculogenesis in the rat, Rattus norvegicus. Serial sections of postnatal ovaries revealed a negligible frequency of follicles 24 h after birth (about 1 primordial follicle per ovary). Twelve hours later there were about 500 follicles per ovary, a number that more than doubled to about 1300 during the subsequent 12 h, indicating that an explosive period of follicular differentiation occurs between the end of postnatal days 1 and 2. Electron microscopy demonstrated that before birth the ovaries are already innervated by fibers containing clear and dense-core vesicles. Immunohistochemistry performed on either fetal (day 19) or newborn (less than 15h after birth) ovaries showed the presence of catecholaminergic nerves, identified by their content of immunoreactive tyrosine hydroxylase (TH), the rate-limiting enzyme in catecholamine biosynthesis. While some of these fibers innervate blood vessels, others are associated with primordial ovarian cells, thereby suggesting their participation in non-vascular functions. Since prefollicular ovaries are insensitive to gonadotropins, the results suggest that the developing ovary becomes subjected to direct neurogenic influences before it acquires responsiveness to gonadotropins. Key words: Ovarian nerves - Development - Folliculogenesis - Tyrosine hydroxylase - Immunohistochemistry - Electron microscopy - Rat (Sprague Dawley)
Numerous reports have appeared describing the intragonadal distribution of ovarian nerves in mammals (for Correspondence to: S. Malamed
reviews see Burden 1978; Baljet and Drucker 1979; Mohsin and Pennefather 1979) and identifying the putative neurotransmitters presumably employed by the innervating fibers to regulate ovarian function (reviewed by Burden 1985; Ojeda etal. 1989; Ojeda and Lara 1989). In spite of this, the developmental phase during which the ovaries become innervated and, more importantly, the temporal and functional relationship that this event may have with the initiation of folliculogenesis (Hirshfield J 991) have not been defined. Resolution of this issue is important because it may provide new insights into potential neuroendorine mechanisms that may contribute to the initiation of follicle formation. Substantial evidence exists that early follicular development is a pituitary-independent phenomenon (Hertz 1963; Peters et al. 1973; Schwartz 1974; Funkenstein et al. 1980), and that responsiveness to gonadotropins is not acquired until the end of the first week of postnatal life (Ben-Or 1963 ; Lamprecht et al. 1973 ; Funkenstein et al. 1980; Sokka and Huhtaniemi 1990). In contrast, the ovary responds to activation of adenylate cyclase (Lamprecht et al. 1973; George and Ojeda 1987; Sokka and Huhtaniemi 1990) and to cyclic AMP itself (George and Ojeda 1987) already during fetal life (George and Ojeda 1987; Sokka and Huhtaniemi 1990). Other studies have demonstrated the ability of vasoactire intestinal peptide (VIP), a putative neurotransmitter/neuromodulator contained in ovarian nerves (Papka etal. 1985; Ahmed etal. 1986; Kannisto etal. 1986), to induce cAMP formation and aromatase activity in fetal ovaries (George and Ojeda 1987). Since gonadotropin receptors are undetectable before postnatal day 4 (Peluso et al. 1976; Siebers et al. 1977; Smith White and Ojeda 1981; Sokka and Huhtaniemi 1990), these observations suggest that the first regulatory influences received by the gland during early development may be neurogenic rather than hormonal. To begin examining this hypothesis it is first required to determine if prefollicular ovaries are innervated. The present study addresses this issue. A preliminary report of these findings has appeared (Malamed et al. 1990).
88
1600 1308:1:217 >, 1 4 0 0 0
1200 r
_r 0
~_ 1 0 0 0
2 800
E 5 0 2 +- 9 9 '-
600
E 400
.~_ LU 2 0 0 1.25+0.75 1.0
1.5 Days after birth
Fig. 1 a-d. Sections of neonatal rat ovaries showing the sharp increase in n u m b e r of primordial follicles that occurs between 24 to 48 h after birth, a Histological aspect of primordial follicles characterized by single layer of flattened somatic cells surrounding each oocyte, • b 24 h ovary, x 4 2 0 ; e 36 h ovary, x 4 2 0 ; d 48 h ovary, x 420. Arrows indicate primordial follicles Fig. 2. Development of primordial follicles in neonatal rat ovary. Columns represent mean values. Vertical lines indicate standard errors of the means, n = 4 for each column Fig. 3a, b, Immunocytochemical localization of tyrosine hydroxylase (TH). a T H immunoreactivity in branching nerve fiber in section of 19-day-old fetal rat ovary. Hilum of ovary in upper left corner; border of ovary in lower right corner. Arrows indicate cell clusters in medullary region, x 210. b TH immunoreactivity in branching nerve fiber in section of rat ovary less than 15 h after birth, x 390
20
89
Fig. 4. Transverse section of nerve fiber (N) encapsulated by ovarian cell in 22-day-old fetal ovary. Open arrows Round and elongate clear vesicles; small arrowhead clear vesicle whose membrane is continuous with plasma membrane of nerve fiber; long arrow paired plasma membrane thickenings without vesicles nearby; large arrowhead region of invagination of ovarian cell plasma membrane which envelops nerve fiber. M Mitochondrion of nerve fiber; m mitochondrion of ovarian cell; O ovarian cell. x 50000. Bar length equals 1 gm Fig. 5. Longitudinal section of two nerve fibers in 22-day-old fetal ovary. Open arrows Round and elongate clear vesicles; short arrows densecore vesicles; long arrows paired plasma membrane thickenings. These membrane appositions couple nerve fiber (N) to nerve fiber and nerve fiber to endothelial cell (E). Thin bars adjacent to bundles of neurofilaments. O Ovarian cell. x 31400. Thick bar: 1 g M
Materials and methods Fetal ovaries were obtained from litters of timed-pregnant Sprague Dawley rats, Rattus norvegicus (Bantin and Kingman, Fremont, Calif.). Postnatal ovaries were from Long-Evans derived rats (bred in Dr. Malamed's laboratory). Sprague Dawley and Long-Evans rats have the same gestation period (20-22 days). The exact postnatal ages of the pups were accurately determined by carefully monitoring the times of birth. The number of follicles per ovary was estimated as follows: The ovaries were first fixed in Gomori 1-2-3 fixative (Humason 1972) at room temperature. Thereafter, they were dehydrated with
ethanol, embedded in paraffin, serially sectioned at 5 gm and stained with hematoxylin and eosin. A primordial follicle was defined as an oocyte surrounded by a single incomplete or complete layer of flattened somatic cells (Fig. 1A). A primary follicle was one with cuboidal follicle cells; few of these appeared and only in 48 h-old ovaries. For 24 (_+ 1) h-old ovaries, counts were made of the numbers of follicles in every section of each ovary. For each 36 ( + 1 ) and 48 (+1) h-old ovary, counts were made of the total numbers of follicles present in every 5th or 6th section of the total number of sections per ovary; the mean total number of follicles per section was then multiplied by the total number of sections of the ovary to obtain the total number of follicles
90 per ovary. Values obtained by this method tend to be spuriously high because the thickness of each section is about one third the mean diameter of a primordial follicle as determined with an objective micrometer and eyepiece graticule. Thus, each primordial follicle appears on more than one section. However, with this method follicles with oocytes that are incompletely surrounded by follicle cells are not counted in sections in which the follicle cells do not appear. This tends to correct the overestimation. Regardless of the validity of the estimated absolute values for numbers of follicles per ovary, the uniform application of the method of calculation to the data from ovaries of all ages examined should produce valid relative values. For electron microscopy, the ovaries were immersed in 50% Karnovsky's fixative (Karnovsky 1965) for 4 h at 4~ C. Post-fixation was in 1% OsO4 in 0.1 M cacodylate buffer, pH 7.4 for 1 h at 0~ C. Uranyl acetate (1% aqueous) for 1 h at room temperature was used for en bloc staining followed by routine ethanol dehydration and embedding in EMbed 812 (Electron Microscopy Sciences, Fort Washington, Pa.). Tissue blocks were sectioned at 70-80 nm thickness and examined in a Philips CM12 electron microscope. For immunohistochemical studies, neonatal rats were anesthetized with ether and perfnsed with 20-25 ml of 0,9% saline solution containing heparin (20 units/ml), 0.008% Lid-o-cain (The Butler Company, Columbus, Ohio) and 0.02% NaNO2 for 3 min. Thereafter, the perfusion was continued at the same flov~ rate with a mixture of 4% paraformaldehyde and 0.1% glutaraldehyde in 0.1 M phosphate buffer, pH 7.4, for 7 rain. The ovaries were excised and immersed in the same fixative for 8 min. They were transferred to 4% paraformaldehyde in 0.1 M phosphate buffer, pH 9.0, for 4 h at 4~ and left overnight in phosphate buffer, pH 7.4, containing 20% sucrose at 4~ after which 20 gm cryostat sections were cut and mounted on poly-l-lysine (Sigma Chemical Company, St. Louis, Mo.) coated slides. Rabbit anti-bovine adrenal tyrosine hydroxylase (TH) (Eugene Tech International, Inc, Allendale, N.J.) at 1 : 500-1 : 5000 dilutions was used for all immunocytochemical studies. For negative controls the primary antiserum was replaced by normal goat serum. Adrenal glands from the same rats that provided the ovaries were used as positive controls. Incubation was overnight at room temperature or, with lower concentrations of antiserum, for 67 h (over the weekend). The ovary sections were prepared according to the Avidin DH: biotinylated horseradish peroxidase H complex (ABC) method (Vector Laboratories, Burlingame, Calif.).
Results At one day after birth ovarian follicles were essentially absent (Figs. 1, 2). In the ensuing 24 h, follicle formation increased sharply to about 500 follicles per ovary at 36 h after birth, and to about 1300 follicles per ovary 12 h later. Fig. I b shows the absence o f follicles 24 h after birth, and the marked increase in follicle number during the next 24 h (Fig. i c, d). Fig. 2 provides the quantitative data for these changes. The follicles present were primordial follicles except for occasional primary follicles ( < 5%) in the 48 h ovaries. The latter had cuboidal follicle cells rather than the flattened follicle cells of the primordial follicle. In contrast to the absence of follicles in newborn ovaries, nerve fibers were detected even during fetal life. Immunocytochemical detection of T H at the light-microscopy level provided evidence for the adrenergic nature of these nerve fibers in the ovary from either fetal (19-day-old, Fig. 3a) or newborn ( < 1 5 h - o l d ) rats (Fig. 3 b) prior to the time when primordial follicles ap-
pear. The TH-labeled fibers observed were in the hilum or medullary part of the ovary; none were found in the cortex. Figs. 4 and 5 show the ultrastructural appearance of nerve fibers identified in a 22-day-old fetal ovary. A cross-section of a fiber suggests that it is enveloped by an ovarian cell (Fig. 4). These early nerve fibers are endowed with dense core and clear vesicles (Figs. 4, 5). Two size populations of pleomorphic clear vesicles were seen. Clear vesicles with a mean diameter of 51.3___ 5.8 nm appear in Fig. 4 and with a mean diameter of 79.3 _+8.2 nm in Fig. 5 (P < 0.01 ; Student's t-test). Some o f these vesicles are elongate; their nominal diameters were calculated by using the means of their long and short dimensions. Dense-core vesicles, most of which are elongate, have a mean diameter of 147 + 14 nm. The mean dimensions of the elongate dense core vesicles are 185_+23 n m x 122_+14 nm. These appear in the same fibers that contain the larger clear vesicles (Fig. 5), but are not co-localized with the smaller clear vesicles (Fig. 4). Paired plasma membrane thickenings were observed; these involve nerve to ovarian cell (Fig. 4), nerve to nerve, and nerve to endothelial cell appositions (Fig. 5). The prenatal nerve fibers appear in the formative ovarian tissue proper (Fig. 4) or are associated with blood vessels (Fig. 5). The frequency of TH-labeled fibers is low; thorough examination by light microscopy of all sections of an ovary revealed up to 6 fibers per ovary. In contrast, T H labeling is abundant in adrenal glands from the same animals that provided the ovaries (not shown). T H labeling was absent in ovary and adrenal gland sections incubated without antibody to TH.
Discussion Although substantial progress has been made towards the understanding of mature ovarian function, little if anything - is known about the regulatory signals that govern early ovarian development. Ovarian function does not become subjected to gonadotropin control until approximately the end of the first postnatal week of life (Lamprecht etal. 1973; Sokka and Huhtaniemi 1990). This refractoriness appears to result mostly from a lack of gonadotropin receptors (Peluso etal. 1976; Siebers et al. 1977; Smith White and Ojeda 1981; Sokka and Huhtaniemi 1990). In other studies (Picon et al. 1985; George and Ojeda 1987), the c A M P analog dibutyryl c A M P was found to increase ovarian aromatase activity even during fetal life, suggesting that the intracellular transduction mechanism that mediates gonadotropin actions in the ovary is intact long before the gonad becomes responsive to either L H or F S H (Picon et al. 1985; George and Ojeda 1987; Sokka and Huhtaniemi 1990). Gonadotropins play a major role in the control of ovarian development and secretory activity, but they are not the only substances that affect ovarian function through activation of cAMP-dependent mechanisms. Norepinephrine and VIP, two neurotransmitters con-
91 tained in ovarian nerves (Burden 1978; Mohsin and Pennefather 1979; Papka et al. 1985; Ahmed et al. 1986; Kannisto et al. 1986; Klein and Burden 1988) stimulate steroidogenesis via the cAMP-mediated transduction pathway (Davoren and Hsueh 1984; Zor and Kliachko 1985). Both have been detected in the neonatal ovary between the second and fifth postnatal day (Ben-Jonathan et al. 1984; Advis et al. 1989), i.e., before the gland acquires responsiveness to gonadotropins. Moreover, VIP was found to stimulate cAMP formation and induce aromatase activity in fetal ovaries, which are unresponsive to LI-I and/or FSH (George and Ojeda 1987). These findings indicate that the immature rat ovary becomes exposed to putative neurotransmitters and develops the capacity to respond to neurotransmitter stimulation during an early, gonadotropin-independent phase of development. Although the ovary may have the intrinsic ability to synthesize catecholamines (Dissen et al. 1990) and VIP (Gozes and Tsafriri 1986), a substantial fraction of these transmitters reaches the ovary via the extrinsic innervation (Lawrence and Burden 1980; Dees et al. 1986). The present study provides ultrastructural evidence that the innervation of the ovary is a fetal event that antedates the postnatal initiation of folliculogenesis. It also demonstrates, by means of immunohistochemistry, that some of the prenatal innervating fibers are catecholaminergic as determined by the presence of immunoreactive TH, the rate-limiting enzyme in catecholamine biosynthesis. The TH-positive fibers are sparse. Thorough light-microscopic examination of all the 20-30 serial sections of each of the 16 ovaries studied immunocytochemically revealed, at most, only six TH-labeled fibers per ovary. However, the same fiber may have appeared in more than one section, and the plane of section may have revealed separate segments of the same nerve fiber. Thus, it is likely that the actual number of nerve fibers per ovary is fewer than six. The existence of ovarian intercellular gap junctions (Burghardt and Anderson 1981) suggests a mechanism by which these initial neural inputs may affect early ovarian function. The electron-microscopic studies revealed the presence of synaptoid appositions of plasma membrane thickenings in the ovarian nerves. There are, however, no vesicles associated with the thickenings suggesting that these paired regions of plasma membranes do not serve for the transmission of nerve impulses (Peters et al. 1991). Also, because the nerve fiber pairings involve an endothelial cell and an ovarian cell as well as another nerve fiber, it is possible that these structures are merely areas of adhesion (puncta adhaerentia). The TH-labeled fibers may be assumed to be efferents. If, as is likely, these autonomic fibers are postganglionic, highly specialized synapses would not be expected. Ovarian nerve terminals resemble other autonomic nerve terminals which are not typically in intimate contact with the cells they influence (Burden 1978; Smolen 1988; Weiss 1988; Nilsson 1983). These terminals (as shown in Fig. 5) may lie within a trough or deep groove formed by the effector cell membrane (Dahl 1970; Fink and Schofield 1971; Nilsson 1983; Malamed et al. 1984). This arrangement
may be an adaptation for targeting the delivery of neurotransmitters to the effector cell. Among the ultrastructural features apparent in the nerve fibers of the fetal ovary are several types of vesicles. Two types of nerve fibers were observed; one type with small, clear, round and elongate vesicles, the other with larger, clear, round and elongate vesicles and large dense-core vesicles. The smaller clear vesicles with a mean diameter of 51 nm were seen in a fiber without dense-core vesicles. The morphological features of the smaller clear vesicles suggest that the fiber containing them is inhibitory, possibly GABAergic (Nilsson 1983; Weiss 1988; Peters et al. 1991). The larger clear vesicles with a mean diameter of 79 nm were in a fiber that also contained round and elongate profiles of dense-core vesicles. The dense-core vesicles (147 nm) observed are most likely peptidergic (Nilsson 1983 ; Weiss 1988 ; Peters et al. 1991). The abrupt initiation of folliculogenesis and the brevity of the interval required for the formation of hundreds of new follicles is of considerable interest and in agreement with recent observations by other authors (Rajah and Hirshfield 1991). The presence of neurotransmitters in pre-follicular ovaries suggested by the present study raises the possibility that the initial formation of follicles is, at least in part, facilitated by signals of neural origin. In addition to direct neural influences, factors related to the development of innervation may contribute to the process of folliculogenesis. For instance, blockade of nerve growth factor (NGF) action at the time of birth has been shown to delay follicular development (Lara et al. 1990). Other reports have described the presence of N G F receptors in thecal cells of ovarian follicles (Dissen et al. 1991) and neurotrophin mRNA expression in granulosa cells and oocytes (Ernfors et al. 1990; Hallbook et al. 1991). These observations raise the intriguing possibility that initiation of folliculogenesis involves the participation of molecules until now thought to be exclusively neurotrophic. Such a role is suggested by recent experiments in which we have found that in vivo or in vitro exposure of neonatal ovaries to antibodies to NGF results in a dramatic reduction in follicular formation (Dissen et al. 1992). Indeed, neurotrophins may be required for the differentiation of other non-neural developing tissues such as kidney tubules (Sariola et al. 1991). The present results do not provide direct evidence in support of either a neural or atrophic factor-dependent mechanism influencing the initiation of folliculogenesis, but conclusively demonstrate that development of innervation precedes both the formation of follicles and the acquisition of responsiveness to gonadotropins at the onset of ovarian maturation.
Acknowledgements. We thank Ms. Janie Gliessman for editorial assistance, and Dr. GeoffreyMcAuliffe,Dr. David Crockett, Mr. Alvin Sharma, Mr. Nadeem Haider, and Ms. Maria E. Costa for technical assistance. This work was supported by NIH Grants HD24870, RR-00163 and HD-18185. This is publication no. 1841 of the Oregon RegionalPrimate Research Center.
92 References Advis JP, Ahmed CE, Ojeda SR (1989) Direct hypothalamic control of vasoactive intestinal peptide (VIP) levets in the developing rat ovary. Brain Res Bull 22:605-610 Ahmed CE, Dees WL, Ojeda SR (1986) The immature rat ovary is innervated by vasoactive intestinal peptide-containing fibers and responds to VIP with steroid secretion. Endocrinology 118:1682-1689 Baljet B, Drukker J (1979) The extrinsic innervation of the abdominal organs in the female rat. Acta Anat 104:243-267 Ben-Jonathan N, Arbogast LA, Rhoades TA, Bahr J (1984) Norepinephrine in the rat ovary: ontogeny and de novo synthesis. Endocrinology 115 : 1426-1431 Ben-Or S (1963) Morphological and functional development of the ovary of the mouse. I. Morphology and histochemistry of the developing ovary in normal conditions and after FSH treatment. J Embryol Exp Morph 11 : 1-11 Burden HW (1978) Ovarian innervation. In: Jones RE (ed) The vertebrate ovary: comparative biology and evolution. Plenum Press, New York, pp 615-638 Burden HW (1985) The adrenergic innervation of mammalian ovaries. In: Ben-Jonathan N, Bahr JM, Weiner RI (eds) Catecholamines as hormone regulators. Raven Press, New York, pp 261-278 Burghardt RC, Anderson E (1981) Hormonal modulation of gap junctions in rat ovarian follicles. Cell Tissue Res 214:181-193 Dahl E (1970) Studies of the fine structure of ovarian interstitial tissue. 3. The innervation of the thecal gland of the domestic fowl. Z Zellforsch 109:212-226 Davoren BJ, Hsueh AJW (1984) Vasoactive intestinal peptide: a novel stimulator of steroidogenesis by cultured rat granulosa cells. Biol Reprod 33 : 37-52 Dees WL, Ahmed CE, Ojeda SR (1986) Substance P (SP) and vasoactive intestinal peptide (VIP)-containingnerve fibers reach the ovary by independent routes. Endocrinology 119:638-641 Dissen GA, Nilaver G, Hill DF, Ojeda SR (1990) Expression of nerve growth factor receptor (NGFrec) and tyrosine hydroxylase (TH) mRNAs in the immature rat ovary (abstract). Soc Neurosci 16:296 Dissen GA, Hill DF, Costa ME, Ma YJ, Ojeda SR (1991) Nerve growth factor receptors in the prepubertal rat ovary. Mol Endocrinol 5 : 1642-1650 Dissen GA, Malamed S, Gibney JA, Hirshfield AN, Ojeda SR (1992) Neurotrophins are required for follicular formation in the mammalian ovary. Soc Neurosci Abstr 18 (in press) Ernfors P, Wetmor C, Olson L, Persson H (1990) Identification of cells in rat brain and peripheral tissues expressing mRNA for members of the nerve growth factor family. Neuron 5:511526 Fink G, Schofield GC (1971) Experimental studies on the innervation of the ovary in cats. J Anat 109 : 115-126 Funkenstein B, Nimrod A, Lindner HR (1980) The development of steroidogenie capability and responsiveness to gonadotropins in cultured neonatal rat ovaries. Endocrinology 106: 98-106 George FW, Ojeda SR (1987) Vasoactive intestinal peptide induces aromatase activity before development of primary follicles and responsiveness to FSH. Proc Natl Acad Sci USA 84:5803-5807 Gozes I, Tsafriri A (1986) Detection of vasoactive intestinal peptide-encoding messenger ribonucleie acid in the rat ovaries. Endocrinology 119: 2606-2610 Hallbook F, Ibanez CF, Persson H (1991) Evolutionary studies of the nerve growth factor family reveal a novel member abundantly expressed in Xenopus ovary. Neuron 6 : 845-858 Hertz R (1963) Pituitary independence of the prepubertal development of the ovary of the rat and the rabbit and its pertinence to hypo-ovarianism in women. In: Grady HG, Smith DE (eds) The ovary. Williams & Wilkins, Baltimore, pp 120 127 Hirshfield AN (1991) Development of follicles in the mammalian ovary. Int Rev Cytol 123:43-101
Humason GL (1972) Animal Tissue Techniques. Freeman, San Francisco, p 15 Kannisto P, Ekblad E, Helm G, Owman C, Sjoberg NO, Stjernquist M, Sundler F, Walles B (1986) Existence and co-existence of peptides in nerves of the mammalian ovary and oviduct demonstrated by immunocytochemistry. Histochemistry 86:25 34 Karnovsky MJ (1965) A formaldehyde-glutaraldehyde fixative of high osmolality for use in electron microscopy. J Cell Biol 27:137 Klein CM, Burden HW (1988) Substance P- and vasoactive intestinal polypeptide (VIP)-immunoreactive nerve fibers in relation to ovarian post-ganglionic perikarya in para- and pre-vertebral ganglia: evidence from combined retrograde tracing and immunocytochemistry. Cell Tissue Res 252:403-410 Lamprecht SA, Zor U, Tsafriri A, Lindner HR (1973) Action of prostaglandin E2 and of luteinizing hormone on ovarian adenylate cyclase, protein kinase and ornithine decarboxylase activity during postnatal development and maturity in the rat. J Endocrinol 57:217-233 Lara HE, McDonald JK, Ojeda SR (1990) Involvement of nerve growth factor in female sexual development. Endocrinology 126:364-375 Lawrence IE Jr, Burden HW (1980) The origin of the extrinsic adrenergic innervation of the ovary. Anat Rec 196:51-59 Malamed S, Gibney JA, Yamasaki DS, Carsia RV (1984) Ultrastructural evidence for adrenocortical-neuronal junctions (abstract). Anat Record 208 : 113A Malamed S, Gibney JA, Ojeda SR (1990) Ovarian innervation develops before initiation of folliculogenesis. Anat Rec 226 : 64A65A Mohsin S, Pennefather JN (1979) The sympathetic innervation of the mammalian ovary. A review of pharmacological and histological studies. Clin Exp Pharmacol Physiol 6:335-354 Nilsson (1983) Autonomic nerve function in the vertebrates. Springer, Berlin, pp 43-44 Ojeda SR, Lara H (1989) The role of the sympathetic nervous system in the regulation of ovarian function. In: Pirke KM, Wuttke W, Schweiger U (eds) The menstrual cycle and its disorders. Springer, Berlin, pp 26-32 Ojeda SR, Lara H, Ahmed CE (1989) The potential relevance of vasoactive intestinal peptide to ovarian physiology. In: Adashi E (ed) Putative intraovarian regulators. Seminars in reproductive endocrinology, vol 7. Thieme, New York, pp 52-60 Papka RE, Cotton JP, Traurig HA (1985) Comparative distribution of neuropeptide tyrosine-, vasoactive intestinal peptide-, substance P-immunoreactive, acetylcholinesterase-positive and noradrenergic nerves in the reproductive tract of the female rat. Cell Tissue Res 242:475-490 Peluso JJ, Steger RW, Hafez ESE (1976) Development of gonadotropin-binding sites in the immature rat ovary. J Reprod Fertil 47 : 55-58 Peters H, Byskov AG, Eaher M (1973) Intraovarian regulation of follicle growth in the immature mouse. In: Peters H (ed) The development and maturation of the ovary and its functions. Excerpta Medica, Amsterdam, pp 20-23 Peters A, Palay SL, Webster HdeF (1991) The Fine Structure of the Nervous System, 3 edn. Oxford University Press, New York, pp 8, 169-186 Picon R, Pelloux MC, Benhaim A, Gloaguen F (1985) Conversion of androgen to estrogen by the rat fetal and neonatal female gonad: effects of dcAMP and FSH. J Steroid Biochem 23 : 9951000 Rajah R, Hirshfield AN (1991) The changing architecture of the rat ovary during the immediate postpartum period: a three dimensional (3D) reconstruction. Biol Reprod 44 [Suppl 1] : 152 Sariola H, Saarma M, Sainio K, Arumae U, Palgi J, Vaahtokari A, Thesleff I, Karavanov A (1991) Dependence of kidney morphogenesis on the expression of nerve growth factor receptor. Science 254:571-573
93 Schwartz NB (1974) The role of FSH and LH and of their antibodies on follicle growth and on ovulation. Biol Reprod 10 : 236-272 Siebers JW, Schmidkke J, Engel W (1977) hCG-insensitivity of the postnatal ovary is due to the lack of a specific receptor. Experientia 33 : 689 Smith White S, Ojeda SR (1981) Changes in ovarian luteinizing hormone (LH) and follicle stimulating hormone (FSH) receptor content and in gonadotropin-induced ornithine decarboxylase activity during prepubertal and pubertal development of the rat. Endocrinology 109:152-161
Smolen AJ (1988) Morp])ology of synapses in the autonomic nervous system. J E1 Micr Tech 10:187 204 Sokk a T, Huhtaniemi I (1990) Ontogeny of gonadotropin receptors and gonadotropin-stimulated cyclic AMP production in the neonatal ovary. J Endocrinol 127:297-330 Weiss L (1988) Cell and tissue, 6th edn. Urban and Schwarzenberg, Baltimore, pp 300-302 Zor U, Kliachko S (1985) The fl-adrenergic system in rat ovarian granulosa cells: hormonal and prostaglandin dependence in viyo and spontaneous development in vitro. In: Ben-Jonathan N I Bahr JM, Weiner RI (eds) Catecholamines as hormone regulators. Raven Press, New York, pp 311-328
