Historical Perspective of Turnover of Dominant Follicles During the Bovine Estrous Cycle: Key Concepts, Studies, Advancements, and Terms

Transcription

1 Historical Perspective of Turnover of Dominant Follicles During the Bovine Estrous Cycle: Key Concepts, Studies, Advancements, and Terms J. J. Ireland,* M. Mihm, E. Austin, M. G. Diskin, and J. F. Roche *Molecular Reproductive Endocrinology Laboratory, Department of Animal Science, Michigan State University Department of Veterinary Preclinical Studies, University of Glasgow Veterinary School, Glasgow, UK Faculty of Veterinary Medicine, University College Dublin, Ballsbridge, Dublin 4, Ireland Teagasc, Athenry, Co. Galway, Ireland ABSTRACT This review chronicles the key concepts, studies, advancements and terms that have led to our current understanding of turnover of dominant follicles (growth and atresia) during the bovine estrous cycle. The twowave concept of follicular development was first proposed in 1960, but remained controversial for the next 28 yr. The concept of the dominant follicle was adapted to cattle in By 1988, ultrasound analysis of individual follicles had demonstrated that heifers usually had two or three distinct waves of turnover of dominant follicles during an estrous cycle. From 1992 to 1993, it was established that a transient rise in serum concentrations of FSH initiated each follicular wave, and a decreased episodic secretion of LH was associated with loss of dominance and the end of a nonovulatory follicular wave. In the past decade, numerous intrafollicular growth factors, such as inhibins, activins, and insulin-like growth factors and their binding proteins, have been identified in follicular fluid of individual bovine follicles. In addition, in vitro studies demonstrate that these growth factors could have endocrine, autocrine, or paracrine actions that modify gonadotropinstimulated follicular growth and differentiation. However, the precise role of intrafollicular growth factors in turnover of dominant follicles has not been defined. We concluded that two or three FSH-stimulated waves of follicular growth usually occur during the bovine estrous cycle, and each follicular wave culminates in development of a single nonovulatory or ovulatory dominant follicle. (Key words: dominant follicles, follicular waves, bovine estrous cycle, key concepts) Abbreviation key: RIA = radioimmunoassay. 1 Received November 12, Accepted March 10, Corresponding author: J. J. Ireland; Ireland@pilot.msu.edu. INTRODUCTION Numerous procedures to artificially manipulate the length of estrous cycles, time of ovulation, or numbers of follicles that grow and ovulate in cattle have been developed (80). However, none have consistently improved fertility or removed the considerable variation in number of follicles or in oocyte quality after superovulation. Further progress in development of more reliable procedures to control follicular development may have been hindered because the physiological concept of a dominant follicle (one that prevents growth of other follicles during the bovine estrous cycle) has only relatively recently been recognized by animal scientists. Indeed, the idea that turnover (growth and atresia) of antral follicles during the bovine estrous cycle is a continuous process without distinct patterns persisted in the literature from 1927 to 1986 (8, 57, 90). However, the development and adaptation of ultrasonography as a noninvasive means to monitor turnover of follicles on consecutive days in the same animal firmly establishes that ovarian antral follicles grow in distinct patterns or waves throughout most of the heifer s life. A large number of excellent recent reviews have addressed the topic of regulation of turnover of dominant follicles in cattle, including the interaction of gonadotropins with intrafollicular growth factors (5, 6, 19, 22, 23, 25, 30, 41, 43, 50, 53, 61, 62, 78, 80, 90, 98, 99). Consequently, the purpose of the present review is to describe, in chronological order, the key concepts, studies, advancements, and terms that have led to our current understanding of turnover of dominant follicles during the bovine estrous cycle. KEY CONCEPTS, STUDIES, ADVANCEMENTS, AND TERMS 1960: The two-wave concept for follicular growth during the bovine estrous cycle is proposed. Animal 2000 J Dairy Sci 83:

2 SYMPOSIUM: FOLLICULOGENESIS IN THE BOVINE OVARY 1649 reproductive scientists were aware before 1960 that preovulatory size follicles existed during the middle of the estrous cycle in cattle (9, 14, 38, 39, 57). In 1960, however, Rajakoski (76) reported results of a 3-yr study of bovine ovaries obtained after slaughter of one to three sexually mature Swedish Red-and-White Breed heifers (usually 1) on each day of the estrous cycle. Follicles 1 mm in diameter were counted, and histological analysis of each sectioned ovary was performed. Based primarily on qualitative analysis, Rajakoski stated that One interpretation for these results is that follicles 5 mm show two growth waves during an oestrus cycle. The first of these occurs between the third and the fourth days and the other between the twelfth and fourteenth. Both result in a follicle of preovulatory size. The large follicle of the first wave undergoes atresia from the twelfth day to leave the large atretic follicle which is present in one of the ovaries from the twelfth to the seventeenth days. The large follicle of the second wave ovulates after final maturation during oestrus (76). Thus, Rajakoski was the first to propose the two-wave concept of follicular turnover during the bovine estrous cycle. Nevertheless, over the next 28 yr, his study remained controversial because investigators reported that their results either supported (37, 44, 67, 93) or refuted the two-wave hypothesis (8, 16, 17, 48, 54, 56, 72, 90) and 1981: Lifespan and fate of individual follicles during the estrous cycle of heifers is directly examined. Dufour et al. (17) and Matton et al. (56) were the first to devise a method to follow changes in size of individual antral follicles on the surface of bovine ovaries. In their studies, groups of heifers from different days of the estrous cycle underwent surgery. They injected dots of India ink at different locations into the stromal tissue surrounding the periphery of either the largest follicle per pair of ovaries (17) or the two largest follicles for each ovary (56). Three to five days after marking, each heifer was subjected to ovariectomy, and the size and fate of each marked follicle was determined. Their results demonstrated that one of the largest follicles on each ovary persists from d 3 to 13 of the estrous cycle, then undergoes atresia (56). This finding directly supported the existence of the first wave of Rajakoski s two-wave hypothesis (76). However, neither of the two largest follicles marked on each ovary between d 13 to 18 of the estrous cycle ovulated (56). Thus, this finding refuted Rajakoski s two-wave hypothesis (76) : Growth and differentiation of estrogen-active and -inactive follicles during the estrous cycle is characterized. Before 1982, the health status of an antral follicle (growing or atretic) in cattle was determined by traditional histological methods (8, 18, 55, 72, 76). In vitro studies in sheep, however, demonstrated that healthy antral follicles produce more estradiol than androgens compared with atretic follicles (7, 63). In contrast, atretic follicles produce more androgens than estradiol compared with healthy follicles (7, 63). Ireland and Roche (47, 48, 49) were the first to classify individual antral follicles ( 6 mm in diameter) in heifers as estrogen-active or estrogen-inactive based on the intrafollicular ratio of concentrations of estradiol to progesterone or androgens in follicular fluid. Estrogen-active follicles (estradiol > progesterone and androgens) have biochemical characteristics of healthy growing follicles, but estrogen-inactive follicles (progesterone or androgens > estradiol) have characteristics of atretic follicles or follicles destined to become atretic (47, 48, 49). In a series of studies (47, 48, 49), ovaries were removed from groups of heifers at 2-d intervals during d 3 to 13 of the estrous cycle, or at different times during a spontaneous or prostaglandin-induced follicular phase. Each follicle ( 6 mm in diameter) per pair of ovaries was classified as estrogen-active or -inactive. Results of these studies established that growth of estrogenactive follicles is associated with an enhanced number of granulosa cells and an increased intrafollicular concentration of estradiol and number of binding sites for LH in granulosa and theca cells both during the follicular phase, and during d 3 to 7 of the luteal phase of the estrous cycle (47, 48, 49). In contrast, the number of binding sites for FSH in granulosa cells decreases. During the follicular phase, but after the preovulatory gonadotropin surge, or during d 9 to 11 of the luteal phase, the largest follicle is estrogen-inactive. Compared with estrogen-active follicles, the estrogen-inactive follicles have a reduced intrafollicular concentration of estradiol and a diminished number of gonadotropin receptor sites in granulosa and theca cells. These studies demonstrate that estrogen-active follicles switch from FSH- to LHresponsiveness as they develop, and that a decreased number of gonadotropin receptors in granulosal and thecal cells are associated with loss of capacity of estrogen-active follicles to produce estradiol. 1984: Ultrasound was used to monitor sizes of follicles during the estrous cycle of heifers. Studies on the dynamics of follicular turnover during the estrous cycle were hindered before 1984 primarily because morphological observations were based on single, point-in-time analysis of ovaries obtained after slaughter (8, 9, 14, 38, 39, 44, 55, 57, 72, 76) or ovariectomy (47, 48, 49) of cattle. In addition, heifers were subjected to surgery at least twice when sequential development of follicles was examined (16, 17, 56, 86). In 1984, two laboratories reported visualization of ovarian structures in a cow (77) or heifers (65) with a linear-array

3 1650 IRELAND ET AL. ultrasound scanner. Pierson and Ginther (65), however, were the first to use ultrasonography to monitor diameter of follicles throughout the estrous cycle of heifers. The reported diameter of a follicle usually refers to the width of the nonechogenic antrum of a follicle, which contains follicular fluid (69, 74). Thus, diameters of follicles subjected to ultrasound are 2 to 3 mm smaller than diameter of the same follicles after dissection (74). Pierson and Ginther (65) characterized follicle growth as follows: 1) growth of a large follicle to an ostensibly ovulatory size followed by regression at approximately mid-cycle, 2) selective accelerated growth of the follicle destined to ovulate approximately 3 d before ovulation, and 3) regression a few days before ovulation of the larger follicles that were not destined to ovulate. These ultrasound results (65), coupled with those of later studies (66, 67, 68), clearly show a bimodal distribution in numbers and sizes of follicles during the estrous cycle of heifers. In a more extensive followup study in heifers, Pierson and Ginther (67) stated that their results support the postulate of two waves of follicular growth during the estrous cycle. 1987: The concept of a dominant follicle as observed in primates is applied to cattle, and the three-wave hypothesis for development of dominant follicles during the estrous cycle is proposed. Hodgen et al. (34) proposed the concept of dominance in 1977 to explain why follicle growth during the primate menstrual cycle is attenuated in the presence of the ovulatory follicle or corpus luteum. Because of marked differences in size between the largest and next largest follicle, two groups postulated that the largest follicle in heifers is dominating or dominant (44, 56). However, the morphological, histological, hormonal, and biochemical evidence supporting the presence of dominant follicles during the bovine estrous cycle was not described until 1987 (43, 50). In reviews (43, 50), it was argued that a single estrogen-active follicle becomes the largest follicle during three different periods of the estrous cycle. In addition, increased serum concentrations of estradiol in one of the two utero-ovarian veins is associated with enhanced intrafollicular concentrations of estradiol in the largest estrogen-active follicle during the same periods of the estrous cycle (24, 46, 47, 48, 49). Based primarily on these results and coupled with the earlier findings from the India-ink marking studies of Dufour et al. (17) and Matton et al. (56), Ireland and Roche (50) concluded that three cycles of development of dominant follicles occur during a bovine estrous cycle. Each cycle of development of a dominant follicle goes through a selection, dominance and atresia or ovulation phase similar to that described for dominant ovulatory follicles during a primate menstrual cycle. 1988: Ultrasound analysis and ovarian maps are used to track growth and atresia of individual follicles throughout the estrous cycle of heifers. In 1988, three different groups of investigators were the first to publish results of four similarly designed ultrasonography studies. Each study used daily ultrasound analysis to monitor growth of individual follicles throughout the estrous cycle of heifers (26, 69, 81, 88). These studies were distinguished from the previous ultrasound reports (65, 66, 67, 68) because location (relative to the anterior and posterior poles, greater curvature and hilus of each ovary, and (or) corpus luteum) and diameter of each follicle 5mm on each ovary were both established during each ultrasound session. In addition, the location and diameter for each follicle was sketched on a diagram or map for each ovary. Results of daily ultrasound analysis and mapping, therefore, established the patterns of growth and regression of each mapped follicle. Of 46 estrous cycles examined for 33 heifers in three separate studies (26, 81, 88), 1 estrous cycle had one follicle wave, 8 cycles had two waves, 35 cycles had three waves, and 2 cycles had four waves of follicular development. These findings show that three waves of dominant follicles per estrous cycle is the most common pattern observed in heifers, which confirmed the three-wave hypothesis (43, 50). In contrast, visual inspection of graphs of ultrasound results for heifers in another study (69) showed predominantly two waves during an estrous cycle, which supported the two-wave hypothesis (76). Taken together, ultrasound analysis shows that cattle usually have two or three waves of follicular development during their estrous cycle. Thus, both the two-wave (76) and three-wave (43, 48, 50) hypotheses are correct. The dynamics of follicular turnover, including maximum diameter, rate of growth or atresia, day of appearance of follicles 5 mm in diameter, and persistence or duration of detection of follicles, were also first described for dominant and nondominant or secondary follicles in the aforementioned studies (26, 81, 88). In brief, the beginning of a wave (also called emergence) is defined as the first day of the estrous cycle a growing follicle 5 mm in diameter in a new wave is detected by ultrasound. With this definition, the average day of the estrous cycle (for heifers with three waves) each wave begins is d 1.9 (range = 1 to 3), 9.4 (8 to 11), and 16.1 (14 to 19) (26, 88). For the first, second, and third waves, the maximum size of each dominant follicle is 12.3 to 15.5, 10.2 to 15.9, and 12.8 to 18.8 mm. Maximum size is reached on d 6 to 6.4, 14.2 to 16, and 21 of the estrous cycle (26, 81, 88). Persistence or duration of detection of a dominant nonovulatory follicle, defined as the interval between its appearance and disappearance, is 11.4 to 17 versus 7.4 to 13.1 d for the first and

4 SYMPOSIUM: FOLLICULOGENESIS IN THE BOVINE OVARY 1651 second wave (26, 81, 88). In contrast, the dominant ovulatory follicle of the third wave persists about 6 d before ovulation (26, 81, 88). The number of nondominant follicles 5 mm for each wave averages 5.3, 3.9, and 4.7, and these follicles persist for about 6 d. The maximum diameter of nondominant follicles is usually 8 mm (26). For heifers that had two waves per estrous cycle, the first and second (ovulatory) waves commence on d 2 to 4 and 10 to 11,and reach maximum diameters of about 13 mm on d 6 and 19, respectively (81, 88). The ovulatory follicle is detected 10 d before ovulation (88). It is not known why there is great variability among heifers (with three follicle waves) in emergence of each wave during an estrous cycle, maximum size of the dominant follicle during each wave, and persistence of a dominant follicle. The following list defines the terms currently used to describe the physiological phenomenon and dynamics associated with turnover of dominant follicles in cattle (Figure 1). Cohort: A group of nearly synchronously growing follicles ([33]; Figure 1). Recruitment: The process whereby a cohort of primordial follicles (primordial follicle = oocyte surrounded by a single layer of squamous pregranulosal cells) begins growth or enters into a trajectory of growth (87), and, thereafter, becomes dependent on gonadotropins for its continued development to ovulatory size ([33], not depicted in Figure 1). After hypophysectomy of laboratory species or sheep, recruitment of primordial follicles and limited growth of a reduced number of preantral follicles occurs (35). This finding implies that gonadotropins are not required for recruitment or early follicle growth. However, gonadotropin treatment of hypophysectomized animals markedly enhances both recruitment and subsequent growth of preantral follicles (35). Thus, gonadotropins are very likely to have an important role both in the recruitment process and rate of follicle growth. Nevertheless, the factor(s) that initiates recruitment has not been defined. Wave: The cyclic pattern of growth of antral follicles. A follicular wave in heifers is usually characterized by growth of a cohort of 24 antral follicles 3 mm in diameter, and atresia of all but a single dominant follicle that continues to grow until it reaches ovulatory size (30). A wave ends in either ovulation or atresia of the dominant follicle. Selection: The process that results in the reduction of the number of follicles in a growing cohort to the species-specific ovulatory quota (33). Selection is complete when the number of healthy follicles in a growing cohort equals the number of follicles that ovulated ((33), Figure 1). As originally defined by Hodgen (42), selection begins coincident with recruitment of a cohort of Figure 1. A model explaining the physiological terms associated with each wave of follicular development during the estrous cycle of heifers. Based on ultrasound analysis, most heifers have one (First wave) or two waves (First wave, Second wave) of follicular development during the luteal phase and a single wave of development (Ovulatory wave) during the follicular phase. Cohort refers to a group of similar sized nearly synchronously growing follicles. Emergence marks the beginning of a wave and is the first day a 4- or 5-mm follicle is the largest in a new wave. The beginning of selection cannot be determined by ultrasonography. However, the end of selection occurs coincident with onset of dominance. Deviation is when growth rates between the dominant and largest subordinate follicle begin to differ. Dominance occurs when the largest follicle in a wave is 1 to 2 mm larger than the next largest follicle and growth of all subordinate follicles ceases. Subordinate follicles are all nondominant follicles in a wave. Loss of dominance marks the end of a wave and is detected at emergence of the next wave. The growing phase for a follicle begins on the day of the estrous cycle of its emergence and ends the day diameter of the follicle ceases to increase. The static phase is from the day follicle diameter ceases to increase (end of growing phase) until the day follicle diameter decreases minus one day. The regressing phase is the last day of the static phase until the follicle is no longer detectable, which is usually when it reaches 4 to 5 mm in diameter. primordial follicles. Thus, ultrasound has insufficient resolution to identify the beginning of a selection phase. However, the end of the selection phase for a wave, which can be detected by ultrasound, occurs coincident with the onset of dominance (defined below). Unraveling the selection process, or more specifically, why one (the selected ) follicle of a cohort develops to ovulatory size as the others undergo atresia is an area of intense research, rich in both elegant models and intriguing hypotheses (5, 6, 19, 22, 23, 25, 30, 33, 35, 36, 41, 43, 50, 61, 62, 78, 80, 83, 99). Emergence: The first day during growth of a cohort of follicles in a wave that the largest follicle detected by ultrasound is 4 (27) or 5 mm in diameter ((26), Figure 1). Thus, emergence, as detected by ultrasound, marks the beginning of a wave. Whether to use 4 or 5 mm as a criterion for emergence depends both on the skill of the ultrasound user and the resolution of the scanner.

5 1652 IRELAND ET AL. Because of variability in time of emergence for the different waves during an estrous cycle, as mentioned earlier, day of emergence is a useful marker for aligning waves among groups of cattle. The interval from emergence of one wave to emergence of the next wave establishs the length of a wave (Figure 1, vertical arrows). Note, however, that ultrasound cannot be used to establish when recruitment begins, as mentioned above. Thus, the precise length of time required for a primordial follicle after its recruitment to become dominant in a wave cannot be established with ultrasound. Dominance: The process whereby a follicle prevents growth of other follicles, or grows in a hormonal milieu unfit for growth of other follicles (42). Based on this definition, onset of dominance is best defined by ultrasound as the first day of the estrous cycle that: 1) the dominant follicle in a wave is at least 1 to 2 mm larger than the next largest follicle, and 2) growth of all subordinate follicles in the same wave ceases ([59], Figure 1, solid lines at top). It may be possible, based on retrospective ultrasound or biochemical analyses, to identify the follicle that is destined to become dominant, but this predestined dominant follicle should not be considered dominant until growth of all subordinates in the wave ceases. Loss or end of dominance (Figure 1, top, arrows) by the dominant follicle of a wave is defined morphologically by ultrasound as occurring on the day of emergence of the subsequent wave. Subordinate follicles: The remaining follicles (usually 2 to 5) in a wave after identification of the dominant follicle by ultrasound (28) (Figure 1). Dynamics of follicular growth: Based on ultrasound determination of diameter, each dominant or subordinate follicle progresses through a growing, static, and regressing period (depicted for dominant follicle in Figure 1, [27]). Some of these periods extend into other waves, especially the regressing phase (Figure 1). Consequently, precise ovarian maps are critical to track turnover of individual follicles and to assign follicles to the appropriate wave. The growing phase for a follicle begins on the day of the estrous cycle of its emergence and ends the day the diameter of the follicle ceases to increase. The static phase is from the day follicle diameter ceases to increase (end of growing phase) until the day follicle diameter decreases minus 1 d. The regressing phase is the last day of the static phase until the follicle is no longer detectable, which is usually when it reaches 4 to 5 mm in diameter (27). Deviation: The divergence of growth rates between the two largest follicles (Figure 1, thick arrows) within a wave, which is retrospectively determined by ultrasound (30). Deviation, however, should not be confused with selection because selection refers to the phenomenon associated with reduction in number of the origi- nally recruited follicles in a wave to the ovulatory quota (33). However, deviation may mark the end of the selection process, especially because the follicle with the greatest growth rate usually emerges first in a wave and becomes dominant (30). In addition, deviation may be caused by one follicle beginning to exert dominance over another. However, whether deviation represents the earliest functional marker for onset of dominance remains to be determined. 1992: A transient peak in basal serum concentrations of FSH preceded each follicular wave and was required for initiation of a wave and onset of dominance. The first radioimmunoassay (RIA) to determine serum concentrations of FSH in cattle was developed and validated by Akbar et al. in 1974 (4). They demonstrated that serum concentrations of FSH peaked coincident with the preovulatory LH surge in heifers. A secondary surge of FSH in heifers was first reported in 1977 by Dobson (15) following RIA of serum samples collected at 2-h intervals for 3 d during the periovulatory period. Her results clearly demonstrated that two transient rises in serum concentrations of FSH occurred in cattle. One significant transient rise of FSH is coincident with the preovulatory LH surge, and the other commences 12 to 24 h after the peak of the gonadotropin surge, as serum concentrations of LH remain relatively low and static. Later studies in cattle using different RIA formats for FSH were confirmatory (73, 79, 94, 96). Shams et al. (86) reported an association of patterns of secretion of FSH with follicular growth in In their study, FSH RIA of serum samples collected at 2- to 6-h intervals was combined with endoscopic examination of ovaries through a permanent fistula at 2-d intervals. They reported that significant FSH peaks occurred on d 4, 8, 12 to 13, 17, and 18 of the estrous cycle, and during the preovulatory LH surge. In addition, they argued that enhanced follicular growth was associated with each FSH peak, although no statistical analysis of their follicular data was reported. As more investigators became familiar with use of ultrasound, and as FSH assays improved, several groups associated the dynamics of turnover of dominant follicles (Figure 1) with changes in patterns of secretion of the gonadotropins and ovarian steroids. Specifically, it was initially demonstrated that the postovulatory rise of FSH preceded the first follicular wave in heifers (94). But, the first report of a significant association of peaks of FSH with follicle waves throughout an estrous cycle was in 1992 by Adams et al. (3). They showed that heifers with two follicular waves also had a significant 50 to 75% increase in serum concentrations of FSH associated with the emergence of each wave. Once concentrations of FSH among heifers are aligned relative

6 SYMPOSIUM: FOLLICULOGENESIS IN THE BOVINE OVARY 1653 Figure 2. A model depicting the relationship of transient peaks of FSH with each wave of follicular development during the estrous cycle of cattle. to wave emergence for each heifer, peaks of FSH were observed on d 1 (day of ovulation) and 10 of the estrous cycle. Each peak of FSH occurred about 1 d before emergence for both the first and ovulatory wave. Although not statistically significant, Adams et al. (3) also reported that heifers with three waves had peaks of FSH on d 1, 9, and 15, approximately 1 d before emergence for each of the three different waves. In subsequent studies by another laboratory, a combination of daily ultrasound and a blood-sampling regimen of 4- to 6-h intervals were used to evaluate the association of serum FSH (11) with follicular waves during the estrous cycle (10, 92). Results of these studies confirmed the results of Adams et al. (3) by showing that a significant transient twofold increase in FSH preceded the first and second wave of follicular development during the luteal phase of the estrous cycle. Despite use of validated RIA for FSH, the majority of laboratories could not reliably distinguish repeatable patterns of FSH during the luteal phase or during pregnancy or postpartum. The best explanation for why significant peaks of FSH are not routinely detected is most likely because emergence, which is associated with peaks of FSH, is variable among heifers (30). Ultrasound, however, makes it possible to align serum concentrations of FSH relative to wave emergence. Thus, a convincing association of a transient rise in FSH with initiation of each follicle wave has been established not only during the estrous cycle ([3, 92], Figure 2), but also in prepubertal heifers (1, 20) and during pregnancy (29) and postpartum anestrous (12). Several laboratories examined the role of FSH in regulation of follicular waves in cattle. In 1979, Miller et al. (60) demonstrated that injections of bovine follicular fluid, a rich source of inhibins (32), during the follicular phase delayed onset of estrus in heifers. Several years later other laboratories reported that injections of bovine follicular fluid delayed follicle emergence (51), or suppressed secretion of FSH without altering secretion of LH in both ovariectomized (45) and intact (73) heifers. Turzillo et al. (94), however, were the first to report in 1990 that the administration of follicular fluid to heifers both suppressed the postovulatory rise in FSH and delayed follicular wave emergence. Results of several subsequent studies were confirmatory (3, 95). More recent studies showed that immunization of heifers against GnRH or infusion of GnRH agonist suppressed both rises of FSH and pulses of LH, and blocked emergence of follicular waves (31, 70, 71). Taken together, these results convincingly demonstrate that a transient rise in FSH is required to stimulate growth of a wave of follicles during the bovine estrous cycle. Although the ascending arm of a transient rise in FSH initiated follicular waves, recent studies show that the subsequent decline in FSH ends selection and stimulates development of the dominant follicle during a wave. Specifically, in 1993 Adams et al. (2) demonstrated that treatment of heifers with FSH at the time of the spontaneous decline of FSH from its peak delays onset of dominance during the first follicular wave. This finding was confirmed several years later by another laboratory (59). Together, these results imply that the descending arm of the postovulatory surges of FSH has an important role in termination of the selection process and stimulation of the onset of dominance (Figure 1). 1993: A decreased episodic pattern of secretion of LH is associated with termination of a follicular wave. Radioimmunoassays to determine serum concentrations of LH were first developed and validated in 1969 by two laboratories (64, 85). Identification of the preovulatory surge of LH coincident with estrus in cattle was initially reported by Shams and Karg (85) in 1969 and Henricks et al. in 1970 (40). However, more than a decade elapsed before Rahe et al. clearly demonstrated (75) that secretion of LH during the estrous cycle of cattle is in episodic patterns. Specifically, heifers were bled via jugular cannulas for 24 h at 10- min intervals on d 3, 10 or 11, and 18 or 19 of an estrous cycle. Results of that study revealed that patterns of LH secretion change from a high frequency of low amplitude pulses during the early luteal phase to a low frequency of high amplitude pulses by midcycle. By estrus, the pattern of episodic LH secretion switches to a high frequency (similar to the early luteal phase) of very high amplitude pulses. Several years later, studies from other laboratories (84, 97) confirmed the luteal phase results of Rahe et al. (75). Results of these earlier studies (75, 84, 97) were then extended by Cupp et al. (13) because they demonstrated that frequency and amplitude of LH pulses are significantly altered on several different days during the luteal phase of the estrous cycle, when progesterone concentrations are high. How-

7 1654 IRELAND ET AL. ever, none of the aforementioned studies examined the relationship of the changing pulse pattern of LH secretion to the dynamics of turnover of dominant follicles during the estrous cycle. Evans et al. (21) were the first to correlate alterations in the dynamics of LH pulses with ultrasound analysis of follicular waves during the estrous cycle of heifers. They demonstrated that frequency of LH pulses decreases, but amplitude increases during growth of the first-wave dominant follicle. In contrast, during loss of dominance of the first-wave dominant follicle and emergence of follicle growth in the second wave, frequency of LH pulses remains unchanged, but amplitude decreases to levels similar to d 3 to 4 of the estrous cycle. Although Cupp et al. (13) did not correlate the alterations in LH pulse frequency and amplitude to follicular waves, their results were similar to those reported by Evans et al. (21). Taken together, these results imply that the pulse pattern of LH secretion varied during a follicular wave during the estrous cycle. Whether alterations in the episodic secretion pattern of LH secretion during the estrous cycle have a role in the regulation of dominant follicle turnover was investigated by several laboratories. In 1990, Sirois and Fortune (89) reported that insertion of a progesterone releasing device intravaginally at midcycle into heifers markedly enhances lifespan, dominance, and estradiol production by dominant follicles after spontaneous luteolysis. In contrast, insertion of two progesterone-releasing devices to mimic luteal phase progesterone levels results in a normal lifespan and turnover of a dominant follicle. Two key studies published in 1993 by Stock et al. (91) and Savio et al. (82), however, demonstrated the important role changes in the dynamics of LH pulses may have on the lifespan of a dominant follicle, as shown in Figure 3. Specifically, endogenous progesterone production decreases in cattle following treatment with prostaglandin F 2α or spontaneous luteal regression. Progesterone releasing intravaginal devices (Figure 3) or norgestomet implants are then inserted to artificially manipulate concentrations of serum progesterone. Ultrasonography is performed daily or every second day of the cycle to monitor development of dominant follicles. Frequent blood-sampling regimens are used throughout the treatment period to characterize the pulse patterns of LH secretion and daily changes in serum concentrations of progesterone. Results of these studies showed that low serum concentrations of progesterone are associated with a higher pulse frequency of LH (Figure 3, dotted line, left panel), and with extended or persistent growth and function of a dominant follicle, as previously shown (89). In contrast, relatively high concentrations of progesterone result in a reduced frequency of LH pulses and loss of dominance of the Figure 3. A model showing the effects of alterations in number of LH pulses on the development of dominant follicles during the estrous cycle of heifers. The corpus luteum (CL) spontaneously regresses or is induced to regress with prostaglandin F 2α. Progesterone releasing intravaginal devices (PRID) are then inserted into a heifer to release either low (first arrow) or higher (2nd arrow in second panel) amounts of progesterone into the blood stream. In turn, number of LH pulses (dotted lines) and lifespan of the dominant follicle are either enhanced (first panel) or decreased (second panel), respectively. dominant follicle (Figure 3, right panel). Both investigators interpreted their results to mean that a high frequency of LH pulses is necessary to maintain dominance, whereas a reduced frequency of LH pulses triggers loss of dominance (Figure 4). Figure 4. A model showing the interrelationships between the patterns of secretion of FSH and LH; recruitment, selection, dominance, and loss of dominance; and the switch from FSH to LH responsiveness of the dominant follicle during the first wave of follicular development for an estrous cycle. Understanding the mechanisms that regulate each phase of development of dominant follicles and elucidating the relationship of those phases to oocyte growth and maturation may be necessary before new more improved methods to regulate ovulation are developed in cattle.

8 SYMPOSIUM: FOLLICULOGENESIS IN THE BOVINE OVARY 1655 Whether the alterations in patterns of episodic secretion of LH during an estrous cycle (13, 21, 75, 84, 97) regulate turnover of dominant follicles remains to be determined. Nevertheless, in support of an important role for alterations in episodic patterns of LH secretion in turnover of dominant follicles, it has been well established that dominant follicles switch from FSH to LH dependency as they develop (47, 48, 49, 100) (Figure 4). Thus, a decrease in LH pulse frequency, such as that observed during the later stages of a follicle wave during an estrous cycle, is not only associated with, but also may cause loss of dominance of the dominant follicle in each follicle wave (Figure 4). The important role of the fluctuating patterns of secretion of both FSH and LH during a follicle wave was recently demonstrated by Gong et al. (31). They used mini-pumps to infuse a relatively low followed by a higher dose of a GnRH agonist in heifers. This treatment regimen results initially in preovulatory LH and FSH surges. These gonadotropin surges are followed by a persistent reduction in serum LH, but normal transient surges of FSH associated with follicular waves. During this period, the dominant follicle of the first wave grows to ovulatory size, then regresses. However, the size of the subsequent dominant follicle in the next wave is reduced to 7 to 9 mm in diameter and persists for several weeks. Infusion of a higher dose of agonist decreased serum FSH, which results in regression of the dominant follicle and abatement of subsequent follicle growth at the 4 mm size. Taken together, these results show the important roles that postovulatory surges of FSH and LH have in initiation of follicular waves, growth of the dominant follicle in each wave to ovulatory size, and perhaps termination of a wave during an estrous cycle present: Intrafollicular factors have potential autocrine or paracrine roles important for dominant follicle turnover. Evidence has accumulated, primarily from in vitro studies over the past decade, that intrafollicular factors, especially growth factors such as inhibin, activin, IGF-I and -II, and their binding proteins, have an important role in regulation of follicular growth, differentiation, and function (5, 6, 19, 22, 23, 25, 30, 41, 43, 50, 52, 53, 61, 62, 78, 80, 90, 98, 99). For example, inhibin and follistatin have negative feedback effects, whereas activin has a positive feedback effect on FSH secretion in many species including cattle (52, 80). In addition to these endocrine roles, inhibin, activin, IGF, follistatin, and IGF binding proteins have autocrine or paracrine actions that stimulate or inhibit function of granulosal and thecal cells (52, 80, 99). Moreover, the action of one factor antagonizes the other, especially inhibin versus activin, activin versus follistatin, and IGF versus IGF binding proteins. Because the aforementioned growth factors and binding proteins are in follicular fluid of bovine follicles (52, 80, 99), alterations in intrafollicular ratios of the various growth factors and their binding proteins may, therefore, have an important role in establishing which follicle in a wave becomes dominant. For example, dominant follicles have lower intrafollicular levels of small molecular weight IGF binding proteins compared with subordinate follicles (80). The reduced level of IGF binding proteins may result in a greater availability of IGF-I, which would enhance gonadotropin action on granulosal and thecal cells within the dominant compared with subordinate follicles. Consequently, differences in intrafollicular ratios of IGF-I:IGF binding proteins could explain why a dominant follicle outgrows and produces more estradiol than subordinate follicles. In support of this idea, we have recently demonstrated in heifers that intrafollicular levels of IGF-binding protein-4 are lower and estradiol concentrations are higher in the first-wave follicle destined to become dominant compared with follicles destined to become subordinate (58). In general, intrafollicular factors are hypothesized to act not only in an endocrine fashion to regulate secretion of gonadotropins, but also in an autocrine or paracrine fashion to modify gonadotropin action, and in turn follicular growth, differentiation, and function. Although this hypothesis has merit, it has been difficult to test in vivo. Thus, the precise mechanisms explaining how gonadotropins interact with intrafollicular growth factors to regulate turnover of a single dominant follicle during a follicle wave remain unclear. SUMMARY, CONCLUSIONS, AND FUTURE QUESTIONS In summary: 1) Antral follicles develop in distinct patterns or waves of 7 to 10 d in length. 2) Two or three waves of follicular growth usually occur during a 21-d estrous cycle. 3) During each wave, a dominant follicle develops to ovulatory size and usually undergoes atresia unless it ovulates. 4) During growth of a cohort of antral follicles in a wave, they undergo recruitment, selection, dominance, and loss of dominance or ovulation. During recruitment, a cohort of primordial follicles begins to grow and all recruited follicles thereafter become dependent on FSH for their continued growth to ovulatory size. Selection reduces the number of recruited follicles in a cohort to the ovulatory quota. Dominance enables a single follicle to prevent growth of other follicles or grow in a hormonal milieu unfit for growth of other follicles. Loss of dominance results in atresia of the dominant follicle, thus initiating growth of a new follicular wave. 5) Based on ultrasound analysis, emergence is the first day a 4- or 5-mm follicle is

9 1656 IRELAND ET AL. the largest in a new wave. Thus, emergence marks the beginning of a wave. Deviation occurs when growth rates between the dominant and largest subordinate follicle begin to differ. The end of selection occurs coincident with onset of dominance. Dominance occurs when the largest follicle is 1 to 2 mm larger in diameter than the next largest follicle and growth of all subordinate follicles ceased. Loss of dominance marks the end of a wave and occurs at emergence of the next wave. 6) The ascending arm of the transient rise in serum FSH initiates a follicle wave, whereas the descending arm ends selection and initiates onset of dominance. 7) During an estrous cycle, a decreased amplitude of LH pulses is associated with loss of dominance of the dominant follicle in a wave. Based on these results, we conclude that: two or three FSH-stimulated waves of follicular growth usually occur during the bovine estrous cycle, and each follicular wave culminates in development of a single nonovulatory or ovulatory dominant follicle. Past studies have firmly established that two or three waves of follicular development occur during the bovine estrous cycles. However, as outlined in Figure 4, the following fundamental questions remain unanswered: How do gonadotropins and intrafollicular factors interact to regulate recruitment, selection, deviation, dominance and loss of dominance during a follicular wave? What causes the great variability among heifers in emergence of each wave, maximum size of the dominant follicle in each wave, and persistence of each wave during an estrous cycle? Do the dynamics (number of follicles, sizes, growth rate) for individual waves influence quality of oocytes? Do factors such as nutrition, aging, parity, or milk yield alter number of follicular waves per estrous cycle? Does number of follicular waves per estrous cycle influence fertility? We speculate that the development of new, more efficient methods to regulate growth of follicles with high quality oocytes to improve fertility in cattle may depend on resolution of the aforementioned questions. ACKNOWLEDGMENTS Research supported by grants to JJI from USDA ( ) and Research Excellence Funds, and JFR from the Department of Agriculture and Food Stimulus Funds. REFERENCES 1 Adams, G. P., A.C.O. Evans, and N. C. Rawlings Follicular waves and circulating gonadotropins in 8-month-old prepubertal heifers. J. Reprod. Fertil. 100: Adams, G. P., K. Kot, C. A. Smith, and O. J. Ginther Selection of a dominant follicle and suppression of follicular growth in heifers. Anim. Reprod. Sci. 30: Adams, G. P., R. I. Matteri, J. P. Kastelic, J.C.H. Ko, and O. J. Ginther Association between surges of follicle-stimulating hormone and the emergence of follicular waves in heifers. J. Reprod. Fertil. 94: Akbar, A. M., L. E. Reichert Jr, T. G. Dunn, C. C. Kaltenbach, and G. D. Niswender Serum levels of follicle-stimulating hormone during the bovine estrous cycle. J. Anim. Sci. 39: Bao, B., and H. A. Garverick Expression of steroidogenic enzyme and gonadotropin receptor genes in bovine follicles during ovarian follicular waves: A review. J. Anim. Sci. 76: Campbell, B.K., R.J. Scaramuzzi, and R. Webb Control of antral follicle development and selection in sheep and cattle. J. Reprod. Fertil. (Suppl.) 49: Carson, R. S., J. K. Findlay, I. J. Clarke, and H. G. Burger Estradiol, testosterone, and androstenedione in ovine follicular fluid during growth and atresia of ovarian follicles. Biol. Reprod. 24: Choudary, J. B., H. T. Gier, and G. B. Marion Cyclic changes in bovine vesicular follicles. J. Anim. Sci. 27: Cole, H. H A study of the mucosa of the genital tract of the cow, with special reference to the cyclic changes. Am. J. Anat. 46: Cooke, D. J., M. A. Crowe, and J. F. Roche Circulating FSH isoform patterns during recurrent increases in FSH throughout the oestrous cycle of heifers. J Reprod. Fertil. 110: Crowe, M. A., V. Padmanabhan, N. Hynes, S. J. Sunderland, W. J. Enright, I. Z. Beitins, and J. F. Roche Validation of a sensitive radioimmunoassay to measure serum follicle-stimulating hormone in cattle: correlation with biological activity. Anim. Reprod. Sci. 48: Crowe, M. A., V. Padmanabhan, M. Mihm, I. Z. Beitins, and J. F. Roche Resumption of follicular waves in beef cows is not associated with periparturient changes in follicle-stimulating hormone heterogeneity despite major changes in steroid and luteinizing hormone concentrations. Biol. Reprod. 58: Cupp, A. S., T. T. Stumpf, F. N. Kojima, L. A. Werth, M. W. Wolfe, M. S. Roberson, R. J. Kittok, and J. E. Kinder Secretion of gonadotrophins change during the luteal phase of the bovine oestrous cycle in the absence of corresponding changes in progesterone or 17β-oestradiol. Anim. Reprod. Sci. 37: Cupps, P. T., R. C. Laben, and S. W. Mead Histology of pituitary, adrenal, and reproductive organs in normal cattle and cattle with lowered reproductive efficiency. Hilgardia 29: Dobson, H Plasma gonadotrophins and oestradiol during oestrus in the cow. J. Reprod. Fertil. 52: Donaldson, L., and W. Hansel Cystic corpora lutea and normal and cystic graafian follicles in the cow. Aust. Vet. J. 44: Dufour, J., H. L. Whitmore, O. J. Ginther, and L. E. Casida Identification of the ovulating follicle by its size on different days of the estrous cycle in heifers. J. Anim. Sci. 34: Erickson, B. H Development and senescence of the postnatal bovine ovary. J. Anim. Sci. 25: Erickson, G. F., and D. R. Danforth Ovarian control of follicle development. Am. J. Obstet. Gynecol. 172: Evans, A.C.O., G. P. Adams, and N. C. Rawlings Endocrine and ovarian follicular changes leading up to the first ovulation in prepubertal heifers. J. Reprod. Fertil. 100: Evans, A.C.O., C. M. Komar, S.-A. Wandji, and J. E. Fortune Changes in androgen secretion and luteinizing hormone pulse amplitude are associated with the recruitment and growth of ovarian follicles during the luteal phase of the bovine estrous cycle. Biol. Reprod. 57: Findlay, J. K An update on the roles of inhibin, activin, and follistatin as local regulators of folliculogenesis. Biol. Reprod. 48:15 23.

10 SYMPOSIUM: FOLLICULOGENESIS IN THE BOVINE OVARY Findlay, J. K Peripheral and local regulators of folliculogenesis. Reprod. Fertil. Dev. 6: Fogwell, R. L., J. L. Cowley, J. A. Wortman, N. K. Ames, and J. J. Ireland Luteal function in cows following destruction of ovarian follicles at midcycle. Theriogenology 23: Fortune, J. E Ovarian follicular growth and development in mammals. Biol. Reprod. 50: Fortune, J. E., J. Sirois, and S. M. Quirk The growth and differentiation of ovarian follicles during the bovine estrous cycle. Theriogenology 29: Ginther, O. J., J. P. Kastelic, and L. Knopf Composition and characteristics of follicular waves during the bovine estrous cycle. Anim. Reprod. Sci. 20: Ginther, O. J., L. Knopf, and J. P. Kastelic Temporal associations among ovarian events in cattle during oestrous cycles with two and three follicular waves. J. Reprod. Fertil. 87: Ginther, O. J., K. Kot, L. J. Kulick, S. J. Martin, and M. C. Wiltbank Relationships between FSH and ovarian follicular waves during the last six months of pregnancy in cattle. J. Reprod. Fertil. 108: Ginther, O. J., M. C. Wiltbank, P. M. Fricke, J. R. Gibbons, and K. Kot Selection of the dominant follicle in cattle. Biol. Reprod. 55: Gong, J. G., B. K. Campbell, T. A. Bramley, C. G. Gutierrez, A. R. Peters, and R. Webb Suppression in the secretion of follicle-stimulating hormone and luteinizing hormone, and ovarian follicle development in heifers continuously infused with a gonadotropin-releasing hormone agonist. Biol. Reprod. 55: Good, T.E.M., J.L.H. Ireland, P.S.D. Weber, J. Pulaski, D. S. McConnell, V. Padmanabhan, A. L. Schneyer, G. Lambert-Messerlian, N. Groome, B. R. Ghosh, W. L. Miller, and J. J. Ireland Isolation of nine different biologically and immunologically active molecular weight variants of bovine inhibin. Biol. Reprod. 53: Goodman, A. L., and G. D. Hodgen The ovarian triad of the primate menstrual cycle. Rec. Prog. Horm. Res. 39: Goodman, A. L., W. E. Nixon, D. K. Johnson, and G. D. Hodgen Regulation of folliculogenesis in the cycling rhesus monkey: selection of the dominant follicle. Endocrinology 100: Greenwald, G. S., and S. K. Roy Follicular development and its control. Pages in The Physiology of Reproduction. 2nd ed. Vol. 1. E. Knobil and J.D. Neill, ed. Raven Press, New York. 36 Greenwald, G. S., and P. F. Terranova Follicular selection and its control. Pages in The Physiology of Reproduction. 1st ed. Vol. 1. E. Knobil and J. D. Neill, ed. Raven Press, New York. 37 Hackett, A. J., and H. D. Hafs Pituitary and hypothalamic endocrine changes during the bovine estrous cycle. J. Anim. Sci. 28: Hammond, J The Physiology of Reproduction in the Cow. Cambridge, Cambridge University Press. 39 Hansel, W The estrous cycle of the cow. Page 232 in Reproduction in Domestic Animals. Vol. 1. H. H. Cole and P. T. Cupps, ed. Academic Press, New York and London. 40 Henricks, D. M., G. D. Dickey, and G. D. Niswender Serum luteinizing hormone and plasma progesterone levels during the estrous cycle and early pregnancy in cows. Biol. Reprod. 2: Hillier, S. G Current concepts of the roles of follicle stimulating hormone and luteinizing hormone in folliculogenesis. Hum. Reprod. 9: Hodgen, G. D The dominant ovarian follicle. Fertil. Steril. 38: Ireland, J.J Control of follicular growth and development. J. Reprod. Fertil. 34: Ireland, J. J., P. B. Coulson, and R. L. Murphree Follicular development during the four stages of the estrous cycle of beef cattle. J. Anim. Sci. 49: Ireland, J. J., A. D. Curato, and J. Wilson Effect of charcoal-treated bovine follicular fluid on secretion of LH and FSH in ovariectomized heifers. J. Anim. Sci. 57: Ireland, J. J., R. L. Fogwell, W. D. Oxender, K. Ames, and J. L. Cowley Production of estradiol by each ovary during the estrous cycle of cows. J. Anim. Sci. 59: Ireland, J. J., and J. F. Roche Development of antral follicles in cattle after prostaglandin-induced luteolysis: changes in serum hormones, steroids in follicular fluid, and gonadotropin receptors. Endocrinology 111: Ireland, J. J., and J. F. Roche Development of nonovulatory antral follicles in heifers: changes in steroids in follicular fluid and receptors for gonadotropins. Endocrinology 112: Ireland, J. J., and J. F. Roche Growth and differentiation of large antral follicles after spontaneous luteolysis in heifers: changes in concentration of hormones in follicular fluid and specific binding of gonadotropins to follicles. J. Anim. Sci. 57: Ireland, J. J., and J. F. Roche Hypotheses regarding development of dominant follicles during a bovine estrous cycle. Pages 1 18 in Follicular Growth and Ovulation Rate in Farm Animals. Vol.. J. F. Roche and D. O Callaghan, eds. Martinus Nijhoff Publishers, The Netherlands. 51 Kastelic, J. P., J.C.H. Ko, and O. J. Ginther Suppression of dominant and subordinate ovarian follicles by a proteinaeous fraction of follicular fluid in heifers. Theriogenology 34: Knight, P. G Roles of inhibins, activins, and follistatin in the female reproductive system. Front. Neuroendocrinol. 17: Lucy, M. C., J. D. Savio, L. Badinga, R. L. De La Sota, and W. W. Thatcher Factors that affect ovarian follicular dynamics in cattle. J. Anim. Sci. 1992: Mariana, J. C., and N. Nguyen Huy Folliculogenese chez la vache. Ann. Biol. Anim. Biochim. Biophys. 13: Marion, G. B., H. T. Gier, and J. B. Choudary Micromorphology of the bovine ovarian follicular system. J. Anim. Sci. 27: Matton, P., V. Adelakoun, Y. Couture, and J. J. Dufour Growth and replacement of the bovine ovarian follicles during the estrous cycle. J. Anim. Sci. 52: McNutt, G. W The corpus luteum of pregnancy in the domestic cow (Bos Taurus) and a brief discussion of cyclical ovarian changes. Am. J. Vet. Med. Assoc. 72: Mihm, M., E. Austin, T.E.M. Good, J.L.H. Ireland, P. Knight, J. F. Roche, and J. J. Ireland Identification of potential intrafollicular factors involved in selection of dominant follicles in heifers. Biol. Reprod. (in Press) 59 Mihm, M., T.E.M. Good, J.L.H. Ireland, J. J. Ireland, P. G. Knight, and J. F. Roche Decline in serum follicle-stimulating hormone concentrations alters key intrafollicular growth factors involved in selection of the dominant follicle in heifers. Biol. Reprod. 57: Miller, K. F., J. K. Critzer, R. F. Rowe, and O. J. Ginther Ovarian effects of bovine follicular fluid treatment in sheep and cattle. Biol. Reprod. 21: Moley, K. H., and J. R. Schreiber Ovarian follicular growth, ovulation and atresia. Adv. Exp. Biol. Med. 377: Monget, P., and D. Monniaux Growth factors and the control of folliculogenesis. J. Reprod. Fertil. (Suppl.) 49: Moor, R. M., M. F. Hay, H. M. Dott, and D. G. Cran Macroscopic identification and steroidogenic function of atretic follicles in sheep. J. Endocrinol. 77: Niswender, G. D., L. E. Reichert, Jr., A. R. Midgley, Jr., and A. V. Nalbandov Radioimmunoassay for bovine and ovine luteinizing hormone. Endocrinology 84: Pierson, R. A., and O. J. Ginther Ultrasonography of the bovine ovary. Theriogenology 21: Pierson, R. A., and O. J. Ginther Ovarian follicular populations during early pregnancy. Theriogenology 26: