Reproduction
(2006)
132
365-377
DOI: 10.1530/rep.1.01233
Copyright © 2006 Society for Reproduction and Fertility
Follicle selection in cattle and horses: role of intrafollicular factors
M A Beg1 and
O J Ginther1,2
1 Department of Animal Health and Biomedical Sciences, University of Wisconsin – Madison, 1656 Linden Drive, Madison, Wisconsin 53706, USA and 2 Eutheria Foundation, Cross Plains, Wisconsin 53528, USA
Correspondence should be addressed to M A Beg; Email: mabeg{at}svm.vetmed.wisc.edu
 |
Abstract
|
|---|
The eminent event in follicle selection during a follicular wave in monovular species is diameter deviation, wherein one follicle continues to grow (developing dominant) and other follicles (subordinates) begin to regress. In cattle, the IGF system, oestradiol and LH receptors are involved in the intrafollicular events initiating deviation as indicated by the following: (1) concentrations of free IGF-I and oestradiol in the follicular fluid and number of LH receptors in the follicular wall increase more dramatically in the future dominant follicle than in the future subordinate follicles before the beginning of deviation and (2) ablation of the largest follicle (LF) or injection of recombinant human IGF (rhIGF)-I into the second LF at the expected beginning of deviation increases the concentrations of oestradiol in second LF before the expected beginning of deviation between second LF and third LF. In horses, an increase in free IGF-I, oestradiol, inhibin-A and activin-A is greater in the future dominant follicle than in other follicles before the beginning of deviation. However, free IGF-I is the only one of these four factors needed for the initiation of deviation in horses as indicated by the following: (1) ablation of LF at the expected beginning of deviation increases the concentrations of free IGF-I in second LF before the beginning of deviation between second LF and third LF but does not increase the other factors; (2) injection of rhIGF-I into second LF at the expected beginning of deviation causes second LF to continue to grow and become a codominant follicle and (3) injection of IGF-binding protein-3 into LF at the expected beginning of deviation causes LF to regress and second LF to become dominant. Thus, the dramatic changes in the IGF system in LF compared to other follicles before the beginning of deviation play a crucial role in the events that lead to the beginning of diameter deviation in both cattle and horses. Oestradiol and LH receptors also play a role in cattle. These intrafollicular events prepare the selected follicle for the decreasing availability of FSH and increasing availability of LH. The other follicles of the wave have the same future capability but do not have adequate time to attain a similar preparatory stage.
|
Introduction
|
|---|
Follicle selection in monovular species is the process wherein only one follicle develops from a wave of growing follicles and continues to grow and ovulates. The eminent ultrasonographically identifiable selection event during a follicular wave is continued growth of the developing dominant follicle and reduced growth of the remaining follicles (subordinates) and is known as diameter deviation (Ginther et al. 1997a). Deviation is preceded by a common growth phase of several days. During this phase, the follicles grow at an approximately similar rate and each follicle has the capacity for future dominance (Gastal et al. 2004). During the common growth phase, no follicle exerts dominance over its cohorts, even though the largest follicle (LF) is most likely to reach a critical development stage first, and become dominant. In this regard, the LF at first detection of a follicular wave became the dominant follicle in about 60% of cattle (Ginther et al. 1996) and horses (Gastal et al. 2004). Therefore, we prefer to define the follicles of the common growth phase on any day according to diameter rank as the LF, second LF, third LF, etc. Definition or use of the terms dominant and subordinate have been inconsistent among investigators. Some laboratories have referred to the two LFs in cattle as dominant and subordinate early in the wave, including at emergence of 3 or 4 mm LF (Evans et al. 2004, Jaiswal et al. 2004, Mihm et al. 2006). In this review, the terms will be used after the beginning of deviation, when follicle destiny is known. For experimental needs, the LF can be defined as the expected dominant follicle when it reaches a diameter, which has been shown to be characteristic of the beginning of deviation in a given species.
Cattle and horses have similar follicle characteristics throughout the ovulatory follicular wave, despite an approximately 2.5-fold greater diameter of follicles in horses (Ginther et al. 2003a). Thus, LF in cattle and horses respectively is about 8.5 and 22.5 mm and second LF is 7.2 and 19.0 mm at the beginning of deviation, and the dominant follicle is about 16 and 40 mm on the day before ovulation. In this regard, the relative follicle diameters throughout the ovulatory wave are also similar between mares and women (Ginther et al. 2004a). In addition, the temporal relationships between follicle deviation and the follicle-stimulating hormone (FSH) profile were similar between mares and women (Ginther et al. 2005a), and on a comparative basis the incorporation of mares into this review is especially appropriate.
The systemic endocrine regulation of folliculogenesis involves the gonadotrophins, but various locally produced hormones and growth factors are involved in regulation of folliculogenesis (Fortune et al. 2004). Before the beginning of diameter deviation, intrafollicular biochemical events ensure future dominance of the selected follicle. The mechanism that triggers the biochemical events is not clear, but it occurs during a progressive decline in circulating FSH concentrations and an early increase in luteinizing hormone (LH). Differences between the two LFs in the concentration of intrafollicular factors at a critical time in the gonadotrophin changes are associated with diameter deviation and apparently underlie a greater responsiveness to the gonadotrophins for the developing dominant follicle than for other follicles (Ginther et al. 2003a, 2004b). The temporal coupling between a change in FSH concentrations and follicular response is close so that the deviation mechanism is established in < 8 h in cattle (Ginther et al. 1999). Thus, after the deviation mechanism is activated, only a short time is needed for interference with the development of the follicles that were selected against.
The intrafollicular factors that are candidates for activation of deviation include those related to the insulin-like growth factor (IGF) system, steroids, inhibin-A/activin-A peptides, gonadotrophin receptors, angiogenic factors and several other intrafollicular factors (Armstrong & Webb 1997, Berisha et al. 2003, Ireland et al. 2004). However, the only factors, which have been temporally or functionally implicated in deviation are IGF-I and its associated system, oestradiol and LH receptors. This review considers the basis and status of this conclusion and the comparative differences between cattle and horses. Recent gene-profiling studies germane to follicle selection are also noted.
|
Experimental approaches
|
|---|
Follicle sampling (Fig. 1A
), primarily of follicular fluid, is the initial and most common approach that has been used to study the mechanism of follicle selection in cattle and horses. Collection of samples has been done using various reference points. These include emergence of a follicular wave at a specified diameter of LF (Evans & Fortune 1997), ovulation (Mihm et al. 2000), oestrus (Austin et al. 2001) and follicle diameter (Rivera & Fortune 2003). Diameters have also been chosen with reference to time of expected deviation (Beg et al. 2001, 2002, Ginther et al. 2002a, 2002b). An approach that has been used in our laboratory is to sample with retrospective reference to the observed beginning of deviation based on a control within each experiment or using a sampling technique that permits continued development of the follicle (Ginther et al. 1997b, Gastal et al. 1999a). The entire amount of follicular fluid has been collected in terminal studies (Beg et al. 2001, 2002, Donadeu & Ginther 2002) or a fraction of the follicular fluid has been sampled in vivo, using transvaginal ultrasound guiding with a small-gauge needle in both cattle (Ginther et al. 1997b) and horses (Gastal et al. 1999a). The in vivo sampling approach seems most appropriate for horses, owing to the large size of the follicles and the internal protective location of the ovarian cortex and therefore the follicle wall. In this species, follicular fluid has been sampled as often as three times from the same follicle at 12 or 24 h intervals (Ginther et al. 2004c, 2004d). In vivo sampling causes a transient reduction in growth rate, but usually with recovery within 24 h. Thereafter, growth is parallel between sampled and nonsampled follicles and follicle status (dominant or subordinate) is maintained, except when a large decrease in diameter occurs as a result of leakage (Gastal et al. 1999a).
View larger version (19K):
[in this window]
[in a new window]
|
Figure 1 Illustration of several approaches for study of the follicle selection mechanism in cattle and horses. LF, largest follicle; DF, dominant follicle; SF, subordinate follicle.
|
|
An approach (Fig. 1B
) involving ablation of LF at an equivalent of the expected beginning of deviation has been used in both cattle (Ko et al. 1991, Adams et al. 1993) and horses (Gastal et al. 1999b, 2004). This approach results in second LF becoming dominant. More detailed studies have shown that this approach induces continued growth and predeviation changes in follicular-fluid factors in second LF (Beg et al. 2002, Ginther et al. 2002a, 2002b). Another approach (Fig. 1C) involves injection of a potential stimulatory factor into second LF to study the effect of the factor on production of other factors and conversion of the second LF to dominant status (Ginther et al. 2004c, 2004d). In cattle IGF-I (Ginther et al. 2004c) and horses (Ginther et al. 2004c, 2004d, 2005b), IGF-I, activin-A, pregnancy-associated plasma protein (PAPP)-A and vascular endothelial growth factor (VEGF) have been used. An approach (Fig. 1D) involving injection of a potential inhibitory substance into LF at the beginning of expected deviation has been used to study the effect of blocking or interfering with a factor on the deviation process. In horses, IGF-binding protein (IGFBP)-3 has been used for this approach (Ginther et al. 2004e). A similar approach in cattle has not been reported.
|
IGF system
|
|---|
The IGF system includes IGF-I and -II, IGFBPs and IGFBP proteases (Spicer 2004). IGF-I stimulates granulosa cell proliferation and synergizes with gonadotrophins to promote differentiation of follicle cells (Spicer & Echternkamp 1995). In vitro effects of IGF-I in cattle include increased proliferation of granulosa cells and oestradiol production (Glister et al. 2001); enhanced sensitivity of granulosa cells to FSH (Monget & Monniaux 1995, Spicer & Echternkamp 1995); increased secretion of inhibin-A, activin-A and follistatin from granulosa cells (Glister et al. 2001) and enhanced LH stimulation of androgen synthesis from theca cells (Stewart et al. 1995).
Results of many studies are compatible with a role for the IGF system in deviation but seem at least partly equivocal, owing to one or more of the following: (1) the reference point was not readily related to the beginning of deviation (Mihm et al. 1997, Spicer et al. 2005); (2) a differential change between LF and second LF began at the equivalent of the beginning of deviation, rather than before the beginning of deviation (Beg et al. 2001, Ginther et al. 2003b) or (3) total IGF-I rather than free IGF-I was considered (De la Sota et al. 1996, Mihm et al. 1997). In cattle, the concentrations of free IGF-I did not increase in the LF in association with deviation, but began to decrease in the second LF before the beginning of deviation (Figs 2 and 3, Beg et al. 2002). Thus, the differential change between the two follicles involved a decrease in second LF rather than an increase in LF. Presumably, a differential decrease in the second LF would favour the LF. In this regard, an increase in free IGF-I in LF before the equivalent of the beginning of deviation has also been reported (Rivera & Fortune 2003). In horses, the concentrations of free IGF-I differentially increased in the future dominant follicle before the beginning of deviation (Fig. 2, Donadeu & Ginther 2002).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 2 Differential changes between largest follicle (LF) vs second LF follicle in concentrations of intrafollicular factors before the beginning of deviation in diameter. DF, dominant follicle; SF, subordinate follicle.
|
|
View larger version (9K):
[in this window]
[in a new window]
|
Figure 3 Mean ± S.E.M. follicle diameters, intrafollicular free IGF-I and oestradiol concentrations and granulosa cell mRNA for LHr of two LFs. The vertical dotted line indicates the beginning of diameter deviation. Differences (P < 0.05) within a follicle and factor are indicated by lower-case letters (a, b). LF, largest follicle. Adapted from Beg et al.(2001, 2002).
|
|
When the LF was ablated in cattle (Beg et al. 2002, Ginther et al. 2002a) at the expected beginning of deviation, diameter of the second LF and concentrations of free IGF-I in second LF were greater in an ablation group than in controls 12 h after ablation (Figs 4 and 5). When LF was ablated in horses (Ginther et al. 2002b), the concentration of free IGF-I in second LF began to increase 12 h before the beginning of deviation between second and third LF (Figs 4 and 6), whereas concentrations of oestradiol, inhibin-A and activin-A increased after the beginning of deviation between second and third LF.
View larger version (20K):
[in this window]
[in a new window]
|
Figure 4 Differences in concentrations of follicular-fluid factors in the second-LF (2nd LF) between controls (intact LF) and ablated group (LF ablated at the expected beginning of deviation). Concentrations were determined 12 h (cattle) and 24 h (horses) after ablation. DF, dominant follicle; SF, subordinate follicle.
|
|
View larger version (19K):
[in this window]
[in a new window]
|
Figure 5 Mean ± S.E.M. concentrations of follicular-fluid factors in the second-LF (2nd LF) in controls (intact LF) and ablated group (LF ablated at expected beginning of deviation) in cattle. Concentrations were determined at 12 h after LF was ablated. Differences (P < 0.05) for each factor between treatment groups are indicated by lower-case letters (a, b). Adapted from Beg et al.(2002).
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Figure 6 Mean ± S.E.M. diameters and follicular-fluid concentrations of free IGF-I and oestradiol in three LFs in horses. The LF was ablated at the expected beginning of deviation, resulting in experimental conversion of 2nd LF and 3rd LF into dominant and largest-subordinate follicles respectively. Experimental diameter deviation began at 24 h (arrow) after ablation, whereas IGF-I and oestradiol began to increase in 2nd LF 12 and 48 h respectively. Adapted from Ginther et al. (2002b).
|
|
The studies on differential concentrations of IGF-I between LF and second LF before the beginning of deviation and before experimental induction of deviation between second LF and third LF in both species strongly support the concept that the IGF-system via IGF-I is an initiator of the beginning of follicle deviation and therefore the selection mechanism.
In cattle, when IGF-I was injected into the stroma of both ovaries 1 day after ovulation, follicular-fluid concentrations of oestradiol increased in small follicles but did not change in large follicles (Spicer et al. 2000); there was no change in follicular dynamics. The role of IGF-I in deviation has been studied by injecting recombinant human (rh)IGF-I into the second LF at the expected beginning of deviation in cattle (Ginther et al. 2004c) and horses (Ginther et al. 2004c, 2004d). In cattle, the exogenous IGF-I increased the follicular-fluid oestradiol concentrations in second LF within 6 h (Fig. 7). The effects on other factors, such as inhibin-A, activin-A or VEGF and on follicle dominance were not studied. In horses, a dose of rhIGF-I that simulated the endogenous concentrations of free IGF-I in the LF was injected into the second LF at the expected beginning of deviation; within 24 h the production of inhibin-A, activin-A and VEGF was stimulated and an increase in androstenedione and IGFBP-2 was prevented. However, oestradiol did not increase until 48 h (Figs 7–9, Ginther et al. 2004d). In addition, intrafollicular injection of a high dose of rhIGF-I into the second LF caused more injected follicles to continue to grow, become dominant, and ovulate than in saline-injected controls (Fig. 10A, Ginther et al. 2004c, 2004d).
View larger version (21K):
[in this window]
[in a new window]
|
Figure 7 Changes in concentrations of follicular-fluid factors in the second-LF (2nd LF) following intrafollicular treatment of the 2nd LF with saline (control group) or rhIGF-I (treated group) at the expected beginning of deviation. LF, largest follicle; DF, dominant follicle; SF, subordinate follicle.
|
|
View larger version (15K):
[in this window]
[in a new window]
|
Figure 8 Schematic summary of positive (+), negative (–), or no detected effect (= ) of an intrafollicular injected factor (circled) on the follicular-fluid concentrations of other factors at 24 h after injection in horses. The second-LF (PAPP-A, IGF-I, activin and VEGF) or LF (IGFBP-3) was treated at the expected beginning of deviation. The arrow indicates the direction of an effect. PAPP-A effects on other factors are presumably exerted through IGF-I. A question mark (?) indicates that the effect was not studied. In cattle, the intrafollicular effect of IGF-I was positive on oestradiol but effects among other factors have not been studied. Adapted from Ginther et al. (2004d, 2004e, 2005b).
|
|
View larger version (20K):
[in this window]
[in a new window]
|
Figure 9 Mean ± S.E.M. concentrations of follicular-fluid factors 12 h (cattle) or 24 h (horses) after intrafollicular treatment of second-LF (2nd LF) with saline (control) or rhIGF-I at the expected beginning of deviation. Oestradiol concentration increased in cattle but not in horses, whereas androstenedione concentration increased in cattle but decreased in horses. Differences (P < 0.05) within an end point and species and between groups are indicated by lower-case letters (a, b). Adapted from Ginther et al. (2004c, 2004d).
|
|
View larger version (11K):
[in this window]
[in a new window]
|
Figure 10 (A) Mean ± S.E.M. diameters of second-LF (2nd LF) in groups treated with a single intrafollicular injection of rhIGF-I or saline at the beginning of deviation in horses. Follicles injected with IGF-I continued to grow and become dominant compared to follicles injected with saline. (B) Mean ± S.E.M. diameter of follicles treated with a single intrafollicular injection of rhIGFBP-3 (LF) at the beginning of deviation. The BP-3 treated LF stopped growing and 2nd LF became dominant. Adapted from Ginther et al. (2004d, 2004e).
|
|
In summary, intrafollicular injection of IGF-I into second LF further indicates that the IGF system via IGF-I is an early participant in the initiation of the deviation mechanism in both cattle and horses.
It has been proposed that IGFBPs exert a pivotal role in the regulation of IGF bioavailability by selectively binding the IGFs and making them unavailable to their receptors (Armstrong & Webb 1997). The IGFBPs are inhibitory to gonadotrophin-induced follicular growth and differentiation and inhibit the actions of IGFs at the level of target cells (Spicer & Echternkamp 1995, Monget et al. 1996). Thus, changes in intrafollicular IGFBPs lead to changes in IGF bioavailability and the up or down regulation of gonadotrophin actions on follicular cells. Four IGFBPs (BP-2, -3, -4 and -5) have been detected in follicular fluid of cattle (Echternkamp et al. 1994, de la Sota et al. 1996, Mihm et al. 2000, Austin et al. 2001) and at least four or five different IGFBPs (BP2, -3, -4, -5 and a high molecular weight complex of 90–135 kDa) in a follicular fluid of horses (Gerard & Monget 1998, Bridges et al. 2002). However, only BP-2, -4 and -5 in cattle (Beg et al. 2001, Rivera & Fortune 2003) and BP-2 in horses (Donadeu & Ginther 2002) have been studied in association with deviation.
In cattle, the levels of BP-4 and -5 were lower in LF apparently before the equivalent of the beginning of deviation (Rivera & Fortune 2003) and thereafter. In horses, free IGF-I concentrations were negatively correlated with BP-2, -4 and -5 levels in follicular and luteal phase small, medium and large follicles (Spicer et al. 2005). In both cattle (Beg et al. 2001) and horses (Donadeu & Ginther 2002), follicular-fluid concentrations of BP-2 were similar between LF and second LF before and at the beginning of deviation (Fig. 2), and concentrations increased in the second LF after the beginning of deviation. Furthermore, ablation of the LF at the expected beginning of deviation was temporally associated with a decrease in BP-2 concentrations in the second LF in cattle (Figs 4 and 5, Beg et al. 2002) but not in horses (Ginther et al. 2002b). In horses, injection of rhIGFBP-3 into the LF at the beginning of expected deviation stopped the growth of LF (Ginther et al. 2004e); the second LF became the dominant follicle (Figs 10B and 11). In addition, the intrafollicular BP-3 injection decreased the follicular-fluid concentrations of free IGF-I, oestradiol, activin-A, inhibin-A and VEGF and increased androstenedione concentrations within 24 h (Figs 8 and 11).
View larger version (20K):
[in this window]
[in a new window]
|
Figure 11 Changes in concentrations of follicular-fluid factors in the LF following intrafollicular treatment of LF with vehicle (control group) or rhIGFBP-3 (treated group) at the expected beginning of deviation in cattle and horses. DF, dominant follicle; SF, subordinate follicle.
|
|
In summary, the results of follicle sampling for BP concentrations in cattle and intrafollicular injection of BP-3 into LF in horses further indicate that the IGF system is involved in the initiation of deviation in both species.
The complexity of the IGF system includes IGFBP proteases, such as PAPP-A that have been described in the follicular fluid of cattle and horses (Mazerbourg et al. 2000, Spicer 2004). The proteases degrade the binding proteins and thus increase the bioavailability of IGF-I in follicles. In cattle, a greater proteolytic activity for BP-4 and -5 occurred in the LF than in the second LF apparently before the equivalent of beginning of deviation and was temporally associated with greater concentrations of free IGF-I in the LF and greater BP-4 and -5 in the second LF (Rivera & Fortune 2003). Proteolytic activity for BP-2 (Mazerbourg et al. 2003), BP-4 (Mazerbourg et al. 2000) and BP-5 (Bridges et al. 2002) has also been reported in dominant follicles but well after the beginning of deviation in horses. In addition, a more recent report (Gerard et al. 2004) indicated a greater BP-2 proteolytic activity in early dominant (25 mm) and late dominant (35 mm) follicles than in subordinate follicles; however, differential proteolytic activity of follicles in relation to deviation has not been reported in horses. In this regard, an intrafollicular injection of PAPP-A into the second LF at the beginning of deviation increased the concentration of free IGF-I in horses (Ginther et al. 2005b). The effects and interrelationships of PAPP-A and IGF-I injection on other follicular-fluid factors are illustrated (Fig. 8). It appears that the induction of proteolytic activity in the IGF system is an early event in the selection of a single dominant follicle. As noted above, free IGF-I decreases in the second LF and remains constant or increases in the LF in cattle, and increases in LF and remains constant in second LF in horses, encompassing the beginning of deviation. Therefore, the second LF may acquire lower BP protease activity in cattle and LF may acquire more such activity in horses; that is, the BP protease activity and the resulting changes in BP concentrations reflect the availability of more IGFs.
The studies on temporality and the effect of treatment with BP proteases further confirm that the IGF system plays an initiating role in follicle selection in both species.
|
Steroids
|
|---|
Oestradiol
A characteristic of the selected follicle is its greater capacity for oestradiol production in both cattle and horses (Ginther et al. 2003a). Once an increase in oestradiol has occurred, it has the capacity to increase its own synthesis further by self-augmenting actions through upregulating thecal synthesis of androgens (Wrathal & Knight 1995) and by increasing pregnenolone synthesis in the granulosa cells and preventing its metabolism to progesterone in both granulosa and theca cells (Fortune & Quirk 1988). Oestradiol has not been shown to have direct effects on follicle growth in horses. However, in other species, it promotes development of preantral follicles and stimulates steroidogenesis in granulosa and theca cells in vitro (cattle), stimulates follicle growth and development in vivo and in vitro and inhibits granulosa cell apoptosis (mice and rats), and increases the sensitivity of granulosa cells to FSH and LH by promoting the expression of their receptors and regulating formation of gap junctions among granulosa cells in vivo (rats; for a review, see Rosenfeld et al. (2001a)). In pigs and sheep, oestradiol promotes the synthesis of IGF-I (Spicer & Chamberlain 2000).
In cattle, follicular-fluid concentrations of oestradiol began to increase differentially in LF vs second LF at (Ginther et al. 1997b, Austin et al. 2001, Beg et al. 2001) or shortly before the expected beginning of deviation (Figs 2 and 3; Mihm et al. 2000, Beg et al. 2002, Ginther et al. 2003b). Similarly, in horses, differential oestradiol increase in the LF began before the beginning of deviation (Fig. 2, Gastal et al. 1999a, Donadeu & Ginther 2002). Following ablation of LF in cattle (Figs 4 and 5, Ginther et al. 2002a) or injection of rhIGF-I into the second LF (Fig. 7, Ginther et al. 2004c) at the expected beginning of deviation, oestradiol increased in second LF before the expected beginning of deviation between second LF and third LF. In contrast, the following similar treatments in horses, oestradiol increased after the beginning of deviation between second LF and third LF (Figs 6 and 7, Ginther et al. 2002b, 2004c, 2004d). In cattle, systemic treatment with antiserum against oestradiol decreased the growth rate of the two LFs and delayed the beginning of deviation independent of the effects on FSH (Beg et al. 2003), further implicating oestradiol in deviation in cattle.
The oestradiol increase in LF before the beginning of deviation plays a role in the initiation of deviation in cattle but not in horses.
|
Androgens and progestins
|
|---|
Androgens and progestins are substrates for oestradiol synthesis in the follicle. Androgens are produced in the theca layer and are then aromatized in the granulosa layer (Fortune & Quirk 1988). Theca and granulosa cells both produce progesterone. There were no differences in androstenedione and testosterone concentrations in the follicular fluid of the two LFs during the period before (Fig. 2) and at the beginning of deviation in both cattle (Beg et al. 2001, 2002) and horses (Donadeu & Ginther 2002). Furthermore, ablation of the LF at the beginning of deviation was not associated with changes in intrafollicular levels of androgens in second LF when it attained the diameter characteristics of beginning of deviation (Fig. 4) in cattle (Beg et al. 2002) and horses (Ginther et al. 2002b). Despite the species similarities associated with the beginning of deviation, a marked species difference occurs after deviation begins; androgen concentrations increase in the dominant follicle and decrease in subordinate follicles in cattle (Stewart et al. 1996, Singh et al. 1998, Beg et al. 2001, 2002), whereas concentrations do not increase in the dominant follicle but do in the subordinate follicles in horses (Donadeu & Ginther 2002). This species difference was further illustrated by the results of intrafollicular injection of IGF-I into the second LF (Fig. 9, Ginther et al. 2004c, 2004d); the concentrations of androgen increased in the treated follicle in cattle but decreased in horses.
Androgens enhance the production of progestins in granulosa cell culture (Fortune & Quirk 1988). Oestradiol also stimulates the synthesis of pregnenolone from granulosa cells. That is, both androgens (directly) and progestins (indirectly) serve as substrates for production of oestradiol in the follicle. In cattle, no differential changes in progesterone concentrations were observed in the two LFs before the beginning of deviation (Beg et al. 2001, 2002). When the LF was ablated at the beginning of deviation in cattle, there was an increase in progesterone concentrations in second LF within 12 h in one study (Beg et al. 2002). However, this increase was not confirmed in a subsequent study (Ginther et al. 2002a). In horses, no differential changes in progesterone concentrations were observed in the two LFs before the beginning of deviation (Donadeu & Ginther 2002). In addition, when the LF was ablated at the beginning of deviation, there was no change in the progesterone concentrations in the second LF (Ginther et al. 2002b).
Considering the lack of a differential change in androgens and progesterone in two LFs before or at the beginning of deviation, it seems unlikely that these factors are involved in the deviation mechanism in either species.
|
Inhibin, activin and follistatin
|
|---|
Inhibins and activins are dimeric glycoproteins consisting of 
and ß subunits, and follistatin is a high affinity activin-binding monomeric glycoprotein. All three hormones are present in follicular fluid of cattle (Knight & Glister 2001) and horses (Donadeu & Ginther 2002). Differences in post-translational processing are responsible for various dimeric and monomeric (free subunits) molecular weight forms of inhibin in follicular fluid. Inhibin enhances the LH-induced androgen production in theca cells of rats and cattle (Hsueh et al. 1987, Wrathal & Knight 1995), and this effect was reduced by activin. Activin induces granulosa cell proliferation; increases FSH receptor expression, granulosa cell steroidogenesis, basal and gonadotrophin-stimulated aromatase activity and oestradiol production; and delays the onset of luteinization and atresia (Knight & Glister 2001). The inhibitory effects of activin on both LH- and oestradiol-induced androgen secretion from theca were reversed by follistatin consistent with its role as an activin-binding protein (Wrathal & Knight 1995). The majority of dimeric forms of inhibins in follicular fluid in cattle are high molecular weight (>160 kDa); the smaller dimeric forms (32–34 kDa) are in low concentrations (Ireland et al. 1994, Austin et al. 2001).
Neither the high molecular weight nor the low molecular weight dimeric forms of inhibin and activin in follicular fluid differentially changed among the three LFs during the selection process in cattle (Austin et al. 2001). Follicular-fluid concentrations of inhibin-A and activin-A in cattle were similar among the three LFs, when LF was a mean of 7.6 mm (Mihm et al. 2000) or when growing from a mean of approximately 5–11 mm (Austin et al. 2001). In a study related to the time of deviation, no differences were found in follicular-fluid concentrations of total inhibin, inhibin-A, activin-A and inhibin-B in the two LFs before the beginning of deviation (Fig. 2; Beg et al. 2002). In another study, ablation of the LF at the beginning of deviation was associated with a transient increase in activin-A in second LF before an increase in oestradiol and IGF-I (Ginther et al. 2002a), but this has not been confirmed. A transient increase in activin-A was not detected in LF or second LF before the beginning of deviation (Ginther et al. 2003b).
In horses, inhibin-A and activin-A concentrations began to increase in the future dominant follicle but not in the future subordinate follicles before the beginning of deviation (Fig. 2), and this difference continued after the beginning of deviation (Donadeu & Ginther 2002). Inhibin-B concentrations did not change before the beginning of deviation. When the LF was ablated at the beginning of deviation, activin-A and inhibin-A concentrations increased in the second LF simultaneously but only after the second LF had attained the diameter characteristic of the beginning of deviation and about 24 h after the increase in IGF-I (Ginther et al. 2002b). In addition, follicular-fluid inhibin-A and activin-A concentrations increased in the second LF and decreased in the LF after an injection of rhIGF-I and rhIGFBP-3 respectively, but not until 24 h after injection (Figs 7, 8 and 11, Ginther et al. 2004d, 2004e). Moreover, an intrafollicular injection of PAPP-A into the second LF at the beginning of deviation increased the concentrations of inhibin-A and follistatin and simultaneously increased the concentrations of free IGF-I within 24 h (Fig. 8, Ginther et al. 2005b).
In conclusion, the experimental results indicate that inhibin-A and activin-A do not play a role in the deviation process in cattle. In horses, despite the differential increase in inhibin-A and activin-A in the LF before the beginning of deviation, the late response in increase in inhibin-A and activin-A in the second LF to ablation of LF and to IGF-I injection into the second LF support a similar conclusion as for cattle.
The intrafollicular ratios of activin:follistatin and activin:inhibin have been suggested to be potentially important parameters regulating folliculogenesis (Glister et al. 2001). Follistatin abolishes the activity of activin-A by binding it and inhibin-A opposes the actions of activin-A. Thus, a ratio of activin-A:follistatin and activin-A:inhibin-A reflects the net amount of unopposed activin (activin tone) likely available for interaction with its receptors. In this respect, FSH and IGF-I increased the activin tone in bovine granulosa cell culture media (Glister et al. 2001). Further, a recent study (Glister et al. 2006) reported a sharp increase in intrafollicular activin tone in 3–6 mm follicles in cattle. Increased activin tone was reflected by a 30-fold increase in activin-A with a concomitant sixfold increase in inhibin-A. Follistatin concentrations exceeded activin-A concentrations until follicles were more than 6 mm. In this regard, a progressive decrease in the follistatin concentrations in cattle was detected in the LF but not others when LF grew from a mean of approximately 5–11 mm (Austin et al. 2001).
Further studies are needed to determine if activin/follistatin plays a role in follicle selection and dominance.
|
VEGF and vascularity
|
|---|
An increase in VEGF, an angiogenic factor, occurred in the follicles of cattle (Berisha et al. 2000) and pigs (Barboni et al. 2000) as diameter increased. The synthesis and secretion of VEGF increased in cultures of granulosa cells in cattle (Schams et al. 2001) and monkeys (Martinez-Chequer et al. 2003) when exposed to IGF-I. VEGF has been shown to stimulate mitosis of endothelial cells and to increase vascular permeability and angiogenesis (reviewed by Redmer & Reynolds (1996), Martinez-Chequer et al.(2003)). An expanded anechoic layer within the wall of the future dominant follicle became apparent 1 day before the beginning of deviation in horses and was attributed to increased vascularization (Gastal et al. 1999c). In horses, follicular-fluid VEGF concentrations were higher in LF than in the second LF the day after the beginning of diameter deviation (Ginther et al. 2004d); however, the earlier temporal relationships before deviation have not been studied. In addition, when rhIGF-I was injected into the second LF at the expected beginning of deviation, VEGF concentrations increased within 24 h and when rhIGFBP-3 was injected into LF, concentrations of VEGF decreased (Figs 7, 8 and 11, Ginther et al. 2004d, 2004e). An intrafollicular injection of VEGF into the second LF at the beginning of deviation increased free IGF-I concentrations and decreased androstenedione but did not affect the concentrations of other factors (Fig. 8; Ginther et al. 2005b).
Follicle-produced VEGF is a candidate for a role in vascular/follicle interrelationships during diameter deviation, but such a role has not been adequately demonstrated.
An increase in vascularity would give the follicle an advantage to receive preferential supply of growth factors, gonadotrophins, steroid precursors and other nutrients required for its continued development. The relationship of follicle vascularity to the beginning of deviation has been studied directly by Doppler ultrasonography in horses (Acosta et al. 2004). Blood flow area began to increase differentially in the future dominant vs subordinate follicle about 1 day before the beginning of diameter deviation. A similar Doppler study in cattle did not show a difference between the two LFs before the beginning of deviation (Acosta et al. 2005). The cause and effect relationships of follicle vascularity and VEGF and the beginning of deviation are not known.
Early increased vascularity in the future dominant follicle has been reported in horses but not in cattle.
|
Receptors
|
|---|
The expression of receptors on ovarian granulosa and theca cells is important for autocrine/paracrine actions of local ovarian factors as well as endocrine actions of systemic hormones. In cattle, oestradiol and progesterone receptors are present in both granulosa and theca cells (Rosenfeld et al. 2001b, Schams & Berisha 2002). The oestradiol receptors were localized in granulosa cells of small (1–5 mm), medium (6–9 mm) and large follicles (>9 mm; Rosenfeld et al. 2001b). Receptor expression for both oestradiol and progesterone was upregulated in dominant follicles (Schams & Berisha 2002), indicating their association with follicle growth and development.
Changes in oestradiol and progesterone receptor expression have not been studied with reference to the beginning of diameter deviation.
A study in cattle found no difference in the expression of IGF-I receptor mRNA between small, medium and large follicles (Armstrong et al. 2000). In contrast, another study found that mRNA expression of IGF-I receptor was higher in granulosa cells from follicles 8 to 10 mm than from follicles 5 to 7 mm (Schams et al. 2002). Furthermore, the expression of IGF-I binding sites in cattle increased from primary to large antral follicles (Wandji et al. 1992).
The expression pattern of IGF-I receptors in relation to deviation is not known.
In cattle, LH-receptor mRNA in granulosa cells was detected in follicles >8 mm (Bao et al. 1997), but not in follicles < 8.0 mm or in subordinate follicles; no difference between follicles in the expression of FSH receptor was found. In contrast, another study found that granulosa cell mRNA expression for LH receptor was minimal in three LFs 2 or 3 days after oestrus (Evans & Fortune 1997), although a temporal relationship to the beginning of deviation was not adequately demonstrated. Another study found that LH-receptor protein in the granulosa cells of the LF was greater on the day after the equivalent of expected beginning of deviation than on the day before the beginning of deviation (Bodensteiner et al. 1996). A recent abstract (Luo et al. 2005), reported that the LH-receptor mRNA was about eight times higher in the LF than in the second LF on a day equivalent to the beginning of deviation. In addition, growth of the largest or dominant follicle, primarily after the beginning of the equivalent of deviation, was associated with an increase in granulosa cell LH-receptor mRNA (Mihm et al. 2006). In a study more specifically related to predeviation, an increase in granulosa-cell LH-receptor mRNA expression was found in LF in cattle at an equivalent of 8 h before the expected beginning of deviation (Beg et al. 2001); expression in second LF did not change (Fig. 3). Therefore, the induction of LH receptors in granulosa cells is one of the early events in selection of a single dominant follicle in cattle. In horses, the LH-receptor protein content in granulosa cells was greater when the follicles were 15–19 mm than in smaller follicles (Goudet et al. 1999), but the results were equivocal in regard to the temporality of differential LH receptor acquisition and the beginning of deviation.
In conclusion, granulosa cells of the LF acquire LH receptors in cattle shortly before the beginning of the deviation but further study is needed in horses.
|
Gene profiling and the selection mechanism
|
|---|
The availability of highly specialized techniques, including suppressive subtraction hybridization and gene array techniques has made it possible to screen and identify genes that are differentially expressed in dominant follicles or subordinate follicles. Gene profiling can be expected to aid in identifying deviation mechanisms at the molecular level. Among monovular farm species, gene profiling during follicle selection has been reported only in cattle (Sisco et al. 2003, Evans et al. 2004, Fayad et al. 2004, Mihm et al. 2006). Using suppression subtractive hybridization, 6 of 22 genes were identified (Sisco et al. 2003). Neither follicle diameter nor follicular-fluid BP-2 level allowed a definitive assignment of future dominance of follicles 1.5 days (before the beginning of deviation) and 2.5 days (apparently near to the beginning of deviation) after ovulation. Apparently higher expression of the gene for aromatase, indicating the potential for oestradiol production, was found in one of the several follicles in each of three heifers 2.5 days after ovulation. This finding indicated that one follicle of the cohort became molecularly distinct for aromatase near the beginning of deviation. In another study (Evans et al. 2004), bovine cDNA microarrays for 53 apoptotic genes were used to screen granulosa and theca cells from LF and second LF in ovaries collected 3 days after emergence of a follicular wave. Eighteen genes were shown to be differentially expressed between the two follicles. Notably, LF had enhanced expression of mRNA for aromatase and LH receptor compared to second LF. In a recent study (Mihm et al. 2006), changes in mRNA abundance of 60 granulosa and 53 theca cell genes were demonstrated in ten cows in the largest growing follicle. The follicles ranged from 7.7 to 16.4 mm, and presumably most follicles were collected after the beginning of deviation. Growth of the follicles was associated with upregulation of expression of genes for LH receptor in granulosa and for TGFß-1-induced antiapoptotic factor in theca cells and downregulation of expression of genes for FSH receptor, inhibin , activin-A receptor type-I and two apoptotic factors in the granulosa cells.
Although the progress in gene profiling is impressive, conclusions from these molecular and genomic studies with regard to follicle selection seem equivocal, owing especially to inadequate reference points for determining the temporal relationships between the genes and the beginning of deviation.
|
Conclusions
|
|---|
The schematic sequence of events during the follicle selection mechanism in cattle and horses is depicted in Fig. 12. In cattle, there is maintenance or an increase in free IGF-I via activation of the IGF system by BP-proteases (PAPP-A) and an increase in oestradiol and LH receptors in LF before the beginning of diameter deviation between LF and second LF. At the same time, the concentrations offree IGF-I decrease in the second LF. In horses, there is a greater increase in several factors in LF than in the second LFat this time, but the IGF system is the only mechanism with a demonstrated positive effect on the beginning of deviation. These conclusions are based on the results of sampling follicles, inducing deviation between second LF and third LF by ablating LF or injecting rhIGF-I into the second LF, or injecting IGFBP-3 into LF. The intrafollicular changes in the future dominant follicle apparently increase the responsiveness of LF to decreasing FSH and increasing LH. The other follicles of the wave have the same capacity for dominance, but do not reach a similar preparatory stage before being negatively affected by the changing gonadotrophin concentrations. Thus, the LF alone continues to grow and becomes dominant.
View larger version (30K):
[in this window]
[in a new window]
|
Figure 12 Schematic model for postulated sequence of events occurring before the beginning of diameter deviation during follicle selection in cattle and horses.
|
|
|
Acknowledgements
|
|---|
This work was supported by the University of Wisconsin Foundation, Madison, WI and the Eutheria Foundation, Cross Plains, WI. We would like to thank the present and past members of our laboratory, who conducted and contributed to the research presented in this review. Thanks are also due to Susan C Jensen for technical assistance. The authors declare that there is no conflict of interest that would prejudice the impartiality of this scientific work.
|
Footnotes
|
|---|
Received 12 April 2006
First decision 26 May 2006
Revised manuscript received 1 June 2006
Accepted 13 June 2006
|
References
|
|---|
Acosta TJ, Gastal EL, Gastal MO, Beg MA & Ginther OJ 2004 Differential blood flow changes between the future dominant and subordinate follicles precede diameter changes during follicle selection in mares. Biology of Reproduction 71 502–507.[Abstract/Free Full Text]
Acosta TJ, Hayashi KG, Matsui M & Miyamoto A 2005 Changes in the follicular vascularity during first follicular wave in lactating cows. Journal of Reproduction and Development 51 273–280.[CrossRef][ISI]
Adams GP, Kot K, Smith CA & Ginther OJ 1993 Effect of a dominant follicle on regression of its subordinates in heifers. Canadian Journal of Animal Science 73 267–275.[ISI]
Armstrong DG & Webb R 1997 Ovarian follicular dominance: the role of intraovarian growth factors and novel proteins. Reviews of Reproduction 2 139–146.[Abstract]
Armstrong DG, Gutierrez CG, Baxter G, Glazyrin AL, Mann GE, Woad KJ, Hogg CO & Webb R 2000 Expression of mRNA encoding IGF-I, IGF-II and type 1 IGF receptor in bovine ovarian follicles. Journal of Endocrinology 165 101–113.[Abstract]
Austin EJ, Mihm M, Evans ACO, Knight PG, Ireland JLH, Ireland JJ & Roche JF 2001 Alterations in intrafollicular regulatory factors and apoptosis during selection of follicles on the first follicular wave of the bovine estrous cycle. Biology of Reproduction 64 839–848.[Abstract/Free Full Text]
Bao B, Garverick HA, Smith GW, Smith MF, Salfen BE & Youngquist RS 1997 Changes in messenger ribonucleic acid (mRNA) encoding luteinizing hormone receptor, cytochrome P450-side chain cleavage, and aromatase are associated with recruitment and selection of bovine ovarian follicles. Biology of Reproduction 56 1158–1168.[Abstract]
Barboni B, Turriani M, Galeati G, Spinaci M, Bacci ML, Forni M & Mattioli M 2000 Vascular endothelial growth factor production in growing pig antral follicles. Biology of Reproduction 63 858–864.[Abstract/Free Full Text]
Beg MA, Bergfelt DR, Kot K, Wiltbank MC & Ginther OJ 2001 Follicular fluid factors and granulosa-cell gene expression associated with follicle deviation in cattle. Biology of Reproduction 64 432–441.[Abstract/Free Full Text]
Beg MA, Bergfelt DR, Kot K & Ginther OJ 2002 Follicle selection in cattle: dynamics of follicular fluid factors during development of follicle dominance. Biology of Reproduction 66 120–126.[Abstract/Free Full Text]
Beg MA, Meira C, Bergfelt DR & Ginther OJ 2003 Role of oestradiol in growth of follicles and follicle deviation in heifers. Reproduction 125 847–854.[Abstract]
Berisha B, Schams D, Kosmann M & Amselgruber W 2000 Expression and localization of vascular endothelial growth factor and basic fibroblast growth factor during final growth of bovine ovarian follicles. Journal of Endocrinology 167 371–382.[Abstract]
Berisha B, Sinowatz F & Schams D 2003 Expression and localization of fibroblast growth factor (FGF) family members during the final growth of bovine ovarian follicles. Molecular Reproduction and Development 67 162–171.[ISI]
Bodensteiner KJ, Wiltbank MC, Bergfelt DR & Ginther OJ 1996 Alterations in follicular estradiol and gonadotropin receptors during development of bovine antral follicles. Theriogenology 45 499–512.[CrossRef][ISI][Medline]
Bridges TS, Davidson TR, Chamberlain CS, Geisert RD & Spicer LJ 2002 Changes in follicular fluid steroids, insulin-like growth factors (IGF) and IGF-binding protein concentration, and proteolytic activity during equine follicular development. Journal of Animal Science 80 179–190.[Abstract/Free Full Text]
De la Sota RL, Simmen FA, Diaz T & Thatcher WW 1996 Insulin-like growth factor system in bovine first wave-dominant and subordinate follicles. Biology of Reproduction 55 803–812.[Abstract]
Donadeu FX & Ginther OJ 2002 Changes in concentrations of follicular fluid factors during follicle selection in mares. Biology of Reproduction 66 1111–1118.[Abstract/Free Full Text]
Echternkamp SE, Howard HJ, Roberts AJ, Grizzle J & Wise T 1994 Relationships among concentrations of steroids, insulin-like growth factor-I, and insulin-like growth factor binding proteins in ovarian follicular fluid of beef cattle. Biology of Reproduction 51 971–981.[Abstract]
Evans ACO & Fortune JE 1997 Selection of the dominant follicle occurs in the absence of the differences in the expression of messenger ribonucleic acid for gonadotropin receptors. Endocrinology 138 2963–297
Top
Abstract
Introduction
