Summary. In Exp. 1, ovulation rate was measured in three groups of Romanov ewes given two injections of 600 i.u. PMSG 3 weeks apart with the ewes intact (Group I, N = 8), a similar treatment with the ewes intact at the first injection and unilaterally ovariectomized at the second (Group II, N = 8), or unstimulated ewes which were hemispayed at the same time as Group II ewes (Group III, N = 6).

In Exp. 2, the follicular population of one ovary was correlated with the number of ovulations induced by 600 i.u. PMSG in the contralateral ovary (10 Romanov ewes).

From 8·4 ± 1·8 (Group I) and 8·2 ± 3·3 (Group II) CL at the first injection, PMSG-induced ovulation rate at the second injection decreased to 3·9 ± 1·8 and 3·7 ± 1·2 in Groups I and II respectively, a value similar for ewes with 1 or 2 ovaries. Furthermore, despite no major changes in the number of antral follicles after the first injection, there was no correlation (r = −0·09) between the response to the two successive injections in intact ewes.

Comparison of the ovarian status of the ovary removed before the PMSG injection (Group II ewes of Exp. 1, ewes of Exp. 2) to the number of CL found in the remaining ovary demonstrated that PMSG-induced ovulation rate was (i) not correlated with the overall antral follicle population (r = 0·62 in Exp. 1, r = 0·49 in Exp. 2), (ii) significantly correlated (r = 0·74, P < 0·05, in Exp. 1; r = 0·85, P < 0·01, in Exp. 2) with the number of healthy follicles 0·8–2·0 mm in diameter, and (iii) negatively correlated with the number of healthy follicles >2 mm in diameter (r = −0·29 in Exp. 1;r = −0·61, P < 0·05 in Exp. 2).

Summary. To investigate the number, size and fate of the follicles > 2 mm in diameter involved in the differentiation of the preovulatory follicle in-vivo monitoring of growth and regression of individual follicles was performed at different times after a PG-induced follicular phase.

In Exp. 1, the 3 largest follicles of each ovary were labelled with ink in the peripheral stroma at 0,4,8,12,24 or 48 h after a PG injection. At a second laparotomy, it was assessed which follicles had ovulated. In Exp. 2, a similar procedure was applied in 12 ewes which underwent repeated laparotomies at 6, 30 and 54 h after PG and after ovulation.

At all times, there was a huge variability in the size at the time of ink labelling of the follicles that finally ovulated ('ovulatory' follicles). A single measurement of follicle size is therefore a poor prediction of the ovulatory follicles. However, a significant reduction in the proportion of non-ovulatory follicles within the size range of ovulatory follicles occurred at 12–24 h after PG, together with a change in follicle hierarchy at 8-24 h after PG.

The results of repeated laparotomies showed that only 60% of follicles > 2 mm at 6 h after PG had grown by 30 h. Selection occurred between 30 and 54 h after PG as indicated by a significant drop in the number of growing follicles to a level at 54 h that equalled the ovulation rate. The growth rate of the ovulatory follicles was 1·4 and 0·4 mm per day between 6 and 30 h and 30 and 54 h after PG respectively. The shrinkage rate of non-ovulatory follicles was 1·6 mm/day.

Summary. The ovaries of new born lambs (15 Ile-de-France and 19 Romanov, 34 ovaries) and of 4-week-old lambs (6 Ile-de-France and 12 Romanov, 18 ovaries) were examined histologically to compare ovarian follicular development in infant lambs of breeds differing in their prolificacy. Breed was the major factor affecting follicular population at birth. Ile-de-France lambs had a higher total number of growing follicles (P < 0·001), and more preantral (P < 0·001) and antral (P < 0·005) follicles than did Romanov lambs. Furthermore, the size of the largest follicles was also reduced in Romanov compared to Ile-de-France lambs. At 4 weeks of age, most of the features of the ovarian follicular population except the mean size of the third largest follicle were similar between the two breeds. However, atresia of antral follicles had appeared only in Ile-de-France and not in Romanov lambs.

When a challenge with exogenous gonadotrophins (1000 i.u. PMSG followed by 1500 i.u. hCG) was attempted, ovulation was triggered in 2/6 and 0/12 Ile-de-France and Romanov lambs respectively. Massive follicular development was noted in 3/6 Ile-de-France lambs but in none of 12 Romanov lambs.

Retardation of follicular development together with retardation in the establishment of ovarian sensitivity to gonadotrophins are therefore features typical of the ovaries of Romanov lambs compared to Ile-de-France lambs during the post-natal period.

Summary. Radioactive microspheres (15 μm diameter) were used to measure capillary blood flow rates in the ovaries and ovarian follicles (Q_f) in high fecund Romanov and low fecund Préalpes-du-Sud ewes at the preovulatory stage of the oestrous cycle. Additionally, assessments of the percentage of arterial blood passing through ovarian arterio-venous anastomoses were obtained.

The mean ± s.e.m. Q_f per unit volume of theca ((ml/min) × 10⁴/mm³) for non-atretic follicles in Romanov ewes was significantly greater (P < 0·05) than that in Préalpes ewes (365·8 ± 42·4, n = 19, compared with 241·3 ± 30·1, n = 14). For each breed, the mean Q_f value for non-atretic follicles was 8–10 times greater than that for atretic follicles.

In Romanov ewes, total Q_f ((ml/min) × 10⁴) and Q_f per unit volume of theca was greatest in small-sized follicles (3· 1–5·0 mm) while in Préalpes ewes, maximum flow was attained in larger-sized follicles (5·1–7·0 mm). The elevated Q_f in small-sized follicles in Romanov ewes may be conducive to more follicles achieving maturation at a smaller diameter in this breed than occurs in the Préalpes ewes.

The absence of flow through ovarian arterio-venous anastomoses in the Romanov, but not in the Préalpes, ewes suggests different mechanisms for controlling the distribution of the total ovarian blood supply in the 2 breeds.

Keywords: follicles; blood flow; sheep; prolificacy

Summary. Plasma FSH concentration was significantly higher in Romanov than Ile-de-France ewe lambs at 5, 6 and 7 weeks of age (P < 0·001, P < 0·02 and P < 0·02, respectively) and at 5, 6 and 7 weeks of age (P < 0·001, P < 0·01 and P < 0·05, respectively) compared to Finn lambs. FSH concentrations were similar and unaffected by time in Ile-de-France and Finn lambs. Ovariectomy at 5 weeks of age produced similar increases in FSH concentrations in Romanov and Ile-de-France ewe lambs, but at 3 months of age the increase in FSH concentrations after ovariectomy was significantly steeper (P < 0·02) in Romanov than Ile-de-France lambs. Sensitivity to oestradiol feedback was related to the age of the lambs. At 5 weeks of age, oestradiol (30 μg in oil per lamb) produced a significant decrease (P < 0·001) in FSH concentrations in Romanov and Ile-de-France lambs, demonstrating that negative feedback can be triggered by oestradiol at this age. Positive feedback after an oestradiol challenge was identified in lambs of both breeds at 6 weeks of age. Sensitivity to the negative feedback of follicular fluid compounds was also established at 5–6 weeks and did not differ between breeds. At 9–10 weeks of age, while there was no breed effect of an oestradiol challenge on FSH concentrations, suppression of FSH concentrations by follicular fluid was shorter in Finn than in the other lambs.

As all the feedback mechanisms are functional at 5–6 weeks of age, it is likely that the between breed differences of FSH profile during infancy are linked to differences in gonadal development.

Keywords: FSH; oestradiol; follicular fluid; lambs; prolificacy

Summary. Despite differences in FSH concentrations ranging from 1·5 ng/ml (Romanov ewes) to 4 ng/ml (Ile-de-France ewes) between the follicular and luteal phases, follicular growth (numbers of follicles growing, growth rates, maximum size reached) was morphologically similar between the two stages of the cycle. Injection of 750 i.u. hCG at Day 6 or 16 of the cycle triggered ovulation of 4·1 ± 0·7 and 4·0 ± 1·3 follicles in Romanov and 2·2 ± 0·5 and 1·7 ± 0·5 follicles in Ile-de-France ewes, respectively, demonstrating that functional differentiation was similar between the two stages of the cycle. As gonadotrophin environment differs between these two stages of the cycle, this suggests that there is a wide flexibility in the amount of gonadotrophins required to trigger terminal follicular growth and that ovarian requirements for gonadotrophins might work through thresholds. When Romanov and Ile-de-France ewes were given similar amounts of exogenous gonadotrophins (1250 i.u. PMSG, 750 i.u. hCG) after hypophysectomy, ovulation rates were close to the usual values (Romanov, 5·5 ± 3·9; Ile-de-France, 1·4 ± 0·5), demonstrating that differences in gonadotrophin concentrations during the follicular phase do not play a major role in the high ovulation of the Romanov compared to the Ile-de-France ewes.

Keywords: FSH; ovulation rate; follicular growth; prolific ewes; stage of the cycle

High prolificacy due to a gene that has a large effect on ovulation rate has been noted in Booroola and Inverdale ewes. High prolificacy in the Belclare breed (a composite developed from stocks selected for very large litter size or high ovulation rate) may be related to the segregation of two genes. The aims of this study were (i) to compare the morphological and functional features of ovulatory follicles from carriers (which could only be heterozygous for the genes of interest) and non-carriers, and (ii) to identify markers of the Belclare genes among secreted or cellular ovarian proteins. Belclare carrier ewes had more ovulatory follicles (4.9 ± 0.4) than did non-carrier ewes (2.0 ± 0.2) (P < 0.001). Ovulatory follicles from carriers were also smaller (4.4 ± 0.1 mm versus 5.7 ± 0.2 mm, P < 0.001) and contained a significantly reduced number of granulosa cells (P < 0.001). However, the proportion of proliferating granulosa cells in ovulatory follicles was similar in both groups. The in vitro secretion of steroids per follicle was only marginally lower in follicles from Belclare carriers compared with non-carriers. Furthermore, similar concentrations of steroidogenic enzymes were present in both groups, indicating that steroidogenic potential per granulosa cell is similar between carriers and non-carriers. Possible markers of the Belclare genes were identified among cellular proteins of follicular walls by two-dimensional PAGE and image analysis. Two spots at 78 and 49 kDa were always absent in samples from non-carriers. When secreted proteins in follicles from carriers were compared with those from non-carriers, two spots at 53 and 41 kDa were restricted to samples from carriers and three spots at 97, 91 and 45 kDa were unique to samples from non-carriers. Interestingly, the spot at 91 kDa is also affected by the Booroola gene.

This experiment was conducted to determine why follicles luteinize faster in the Meishan breed than in the Large White breed of pig. Follicles were recovered during the late follicular phase from ovaries of both breeds before and after administration of hCG given to mimic the LH surge. First, the patterns of cholesterol transporters (high and low density lipoproteins: HDL and LDL) were compared. Cholesterol transporters detected in follicular fluid consisted of HDL only. Similar amounts of Apolipoprotein A-I were found in all samples. There was no obvious breed effect on minor lipoproteins found in the HDL-rich fraction, and this pattern was altered similarly by hCG in the two breeds. The LDL-rich samples of serum from both breeds contained similar amounts of protein. Second, three steroidogenic enzymes, adrenodoxin, 17α-hydroxylase–lyase (P450_17α) and 3β-hydroxysteroid-dehydrogenase (3β-HSD) were detected by immunohistochemistry and quantified by image analysis on sections of the two largest follicles. Before hCG treatment, theca interna cells demonstrated immunoreactivities for adrenodoxin (strong), P450_17α and 3β-HSD (very strong), whereas granulosa cells displayed immunoreactivities for adrenodoxin only. After hCG treatment, the localization of the enzymes was unchanged but the staining intensity of adrenodoxin on granulosa cells and 3β-HSD on theca cells increased (P < 0.01 and P < 0.05, respectively). Breed effects were detected for the amounts of adrenoxin in theca cells (Meishan > Large White; P < 0.05) and of 17α-hydroxylase (Large White > Meishan, P < 0.01). Breed × treatment interactions were never detected. Finally, gelatinases, plasminogen activator, plasminogen activator inhibitor, tissue inhibitors of metalloproteases (TIMP-1 and TIMP-2) were visualized by direct or reverse zymography or western blotting. Whatever the stage relative to LH administration, follicular fluid from Large White gilts contained more TIMP-1, and TIMP-2 (P < 0.02 and P < 0.01, respectively). No breed effect was detected for the amounts of gelatinases and plasminogen activator inhibitor 1. However, for these parameters, a significant breed × time interaction was obvious, as the Meishan follicles had a greater response to hCG (P < 0.01). Since proteolysis plays a key role in the bioavailability of growth factors such as insulin-like growth factor 1, fibroblast growth factor and transforming growth factor β, which have the ability to alter gonadotrophin-induced progesterone production in pigs, the differences observed in its control in the present study may explain, at least in part, the different patterns of luteinization observed in Meishan and Large White follicles.

Summary. Prolific breeds of sheep (Romanov, Finn and Booroola Romanov crosses heterozygous for the Booroola gene (F +) were compared with breeds of lower prolificacy (Ile-de-France, Finn × Scottish Blackface, Merino × Blackface and Booroola × Romanov not carrying a copy of Booroola gene (++)) by in-vivo monitoring of follicular kinetics by ink labelling during the late luteal phase and follicular phase of the oestrous cycle followed by histological examination of the ovaries or follicle dissection. At each of 3 successive laparotomies, the 3 largest follicles of each ovary were measured and ink labelled. At the final laparotomy, around the beginning of oestrus, all ewes were ovariectomized.

High ovulation rate was not associated with the total number of antral follicles in any of the breeds. However, there were more follicles > 2 mm in diameter in Romanov and Booroola × Romanov crosses (F +) compared to their respective controls. Such a feature was not observed in Finnish Landrace compared to Finn × Blackface and Merino × Blackface ewes. A more numerous population of recruitable follicles, together with a similar incidence of selection through atresia, were the features associated with the high ovulation rate of Romanov compared to Ile-de-France ewes. The high ovulatory potential of the Finn ewes resulted from a markedly reduced incidence of selection through atresia. Booroola × Romanov ewes carrying a copy of the Booroola gene (F +) appeared to possess features of both parental breeds, including high numbers of recruitable follicles, smaller follicular size when recruitment occurs and an extended time for recruitment. Booroola × Romanov (++) ewes, not carrying the gene, appeared to have lost part of the 'Romanov characteristics' of a more numerous population of recruitable follicles. The variability in the kinetics of preovulatory enlargement, seen in these breeds of sheep, demonstrates that there are a number of pathways through which high ovulation rate can be achieved and hence through which ovulation rate might be manipulated.

Summary. To investigate the factors contributing to the different ovulation rates observed in two strains of sheep (Booroola 5·2, Merino 1·2), in-vivo monitoring of follicular kinetics followed by histological examination of both ovaries was performed during the late luteal and follicular phases. Ewes of both strains were either ovariectomized at Day 13, or had the 3 largest follicles of each ovary ink-labelled at Day 13 and were ovariectomized at Day 15, or had the 3 largest follicles of each ovary ink-labelled at Days 13 and 15 and were ovariectomized 16 h after the beginning of oestrus (N = 6 per time per strain). In another experiment, the age effects on the follicular populations of these two strains were also studied.

There were 2–4 times more primordial follicles and 1·–2 times more preantral follicles in the ovaries of Booroola than in control Merino ewes, although the number of antral follicles was the same. The percentage of normal follicles in this population was higher in Merino than Booroola ovaries. In Booroola ewes, there was no correlation between the number of antral follicles per ovary and the ovulation rate at the previous cycle (r = 0·22). This suggests that follicle numbers do not play a key role in the high ovulation rate of the Booroola strain. The number of follicles initiating growth from the primordial pool, the number of growing follicles disappearing at the preantral stage, the pattern of antrum development, granulosa cell multiplication and appearance of atresia differed between strains.

The reasons for the high ovulation rate of the Booroola strain became clear when preovulatory enlargement was followed by ink labelling. An extended period of time during which recruitment of ovulatory follicles takes place, together with a low incidence of selection and the ability of the follicles to wait for ovulation are the features involved in this high ovulation rate.