- Email: [email protected]
Ram novelty and the duration of ram exposure affects the distribution of mating in ewes exposed to rams during the transition into the breeding season
Animal Reproduction Science 111 (2009) 249–260
Ram novelty and the duration of ram exposure affects the distribution of mating in ewes exposed to rams during the transition into the breeding season P.A.R. Hawken ∗ , A.P. Beard School of Agriculture, Food and Rural Development, Newcastle University, Newcastle Upon Tyne NE1 7RU, United Kingdom Received 4 October 2007; received in revised form 27 February 2008; accepted 14 March 2008 Available online 20 March 2008
Abstract This study compared the affect of short-term and continuous exposure to rams during the transition between anoestrus and the breeding season on the distribution of mating and subsequent lambing. Further, within ewes continuously exposed to rams we investigated the effect of replacing these rams every 17 days with ‘novel’ rams. During August (late anoestrus, Northern Hemisphere), multiparous, North of England mule ewes were allocated to one of four groups: SVR ewes were exposed to vasectomised rams for 24 h on Day 0 (short term; n = 109), RVR ewes were exposed to vasectomised rams for 24 h on Days 0, 17 and 34 (short term; n = 113); PVR ewes were exposed to vasectomised rams on Day 0 and remained with the same rams for the duration of the pre-mating period (continuous; n = 104); NVR ewes were continuously exposed to vasectomised rams from Day 0 with the rams replaced with ‘novel’ rams every 17 days (continuous; n = 113). Blood samples were collected from a subset of ewes (n = 22 per group) to monitor progesterone. On Day 50, harnessed, entire rams were introduced for mating and raddle marks recorded daily for the first 17 days. The median date of mating occurred 1 day earlier in NVR ewes than PVR ewes (P < 0.05). A synchrony score calculated from the blood sampled ewes showed that the distribution of mating was more synchronised in PVR and NVR ewes than SVR and RVR ewes (P < 0.001). PVR and NVR ewes had an earlier onset of cyclic activity than RVR ewes (P < 0.01). However, only NVR ewes differed from SVR ewes in this variable (P < 0.05). Within ewes lambing to first service, the median date of lambing of PVR, NVR and SVR ewes occurred at least 2 days earlier than RVR ewes (at least P < 0.05). Further, PVR and NVR ewes had a more compact distribution of lambing than SVR and RVR ewes (P < 0.05) and lambing was more compact in NVR ewes than PVR ewes (P < 0.05). In conclusion, ewes in continuous contact with rams
∗ Corresponding author. Present address: M085, School of Animal Biology, Faculty of Natural and Agricultural Sciences, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. Tel.: +61 8 6488 3588; fax: +61 8 6488 1029. E-mail address: [email protected] (P.A.R. Hawken).
0378-4320/$ – see front matter © 2008 Published by Elsevier B.V. doi:10.1016/j.anireprosci.2008.03.009
250
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
prior to mating had a more synchronised distribution of mating and lambing than ewes given only short-term exposure to rams. This distribution of mating in continuous ram exposed ewes can be further enhanced by periodic exposure to novel rams. © 2008 Published by Elsevier B.V. Keywords: Ram effect; Breeding season; Novelty; Anoestrus; Ewes
1. Introduction The ram effect is the physiological response of ewes to rams and is a well-established method of stimulating ovulation and oestrus in anovulatory ewes (reviews; Martin et al., 1986; Rosa and Bryant, 2002; Ungerfeld et al., 2004). However, its use in genotypes native to temperate regions is restricted by breed seasonality (Lindsay and Signoret, 1980). Within these highly seasonal breeds, the proportion of ewes ovulating in response to rams only reaches acceptable levels close to the onset of the natural breeding season (Rosa and Bryant, 2002). Therefore, though the ram effect is relatively ineffective at stimulating ovulation during anoestrus in these breeds, rams can be used to control the onset of cyclic activity and the distribution of mating during the breeding season itself (Hawken et al., 2007a). In our previous study, we found that repeated, short-term exposure of North of England mule ewes to rams during the transition into the breeding season compacted the distribution of mating compared with ewes isolated from rams prior to mating (Hawken et al., 2007a). However, this premating strategy may be improved by leaving the rams with the ewes until mating, as the ovulatory response to the ram effect is positively correlated with the duration of ram presence (Signoret et al., 1982). One potential problem with this theory is that ewes maintained in continuous ram contact can become habituated or refractory to the male stimulus (Martin et al., 1986). However, more recent studies in sheep and goats indicate that habituation to male presence can be overcome by frequent replacement of males with unfamiliar or ‘novel’ males (Cushwa et al., 1992; Veliz et al., 2006). In this study we compared the effect of continuous and short-term exposure to rams during the transition between anoestrus and the breeding season on the distribution of mating later during the breeding season and subsequent lambing. Furthermore, in ewes continuously exposed to rams, we tested if replacement of these rams with ‘novel’ rams every 17 days would affect the distribution of mating and lambing. 2. Materials and methods 2.1. Animals and experimental procedures The experiment was carried out in accordance with the Animals (Scientific Procedures) Act 1986 and was approved by the University of Newcastle Animal Ethics committee and the UK Home Office. During August (late anoestrus, Northern Hemisphere) multiparous, North of England mule ewes (Scottish Blackface × Bluefaced Leicester) that had been previously isolated from rams (not within 500 m for a minimum of 2 months) were allocated to either short-term or continuous ram exposure groups. The groups which tested the efficacy of short-term exposure to rams were
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
251
single ram exposure ewes (SVR; n = 109) and repeated ram exposure ewes (RVR; n = 113). The groups that tested the efficacy of continuous ram exposure were permanent ram exposure ewes (PVR; n = 104) and novel ram exposure ewes (NVR; n = 113). All groups were balanced for age (4–6 years) and maintained on pasture at Cockle Park Farm, Northumberland, UK (55◦ 13 N). 2.2. Pre-mating period SVR and RVR ewes were exposed to harnessed, vasectomised rams in the intermittent exposure paddock (0.6 ha) for 24 h on Day 0 of the experiment (11 September). SVR ewes were subsequently isolated from rams for the remainder of the pre-mating period in a paddock more than 500 m from the exposure paddock. RVR ewes were also exposed to harnessed, vasectomised rams for 24 h on Days 17 and 34 of the experiment. Between each 24 h exposure period, the RVR ewes were isolated from rams in a paddock more than 500 m from the exposure paddock. To provide a degree of ram novelty during each exposure, the RVR ewes had not previously been exposed to one of the rams; the other two vasectomised rams remained the same for all three exposure periods. PVR ewes were exposed to harnessed, vasectomised rams on Day 0. The ewes remained with these rams for the duration of the pre-mating period. NVR ewes were exposed to harnessed, vasectomised rams on Day 0. The ewes remained with these rams for the first 17 days until they were replaced with ‘novel’ rams on Days 17 and 34 of the experiment. All paddocks were between 4.5 and 5.0 ha and sown with a comparable pasture mix (rye grass/clover mix). No direct measurements of pasture growth or quality were taken, but visual assessment of the paddocks indicated comparable food availability between paddocks throughout the experiment. The paddocks where rams were permanently present (PVR and NVR) were 6 km from the intermittent exposure paddocks to prevent any impact of these rams on SVR and RVR ewes. The vasectomised rams were crossbred (Scottish Blackface or Suffolk × Border Leicester; n = 3 per exposure) and sexually experienced (i.e. at least 2 years mating experience). They were fitted with harnesses to allow detection of oestrus and were in direct physical contact with the ewes. Raddle marks were recorded by a single observer after each 24 h exposure period in the SVR and RVR ewes and twice weekly in the PVR and NVR ewes. The raddle colour was changed before each 24 h exposure period of RVR ewes, on Days 17 and 34 in the PVR ewes and with each set of novel rams in the NVR ewes. When the vasectomised rams were not being used to stimulate ewes, they were kept with other rams and isolated from ewes (i.e. more than 500 m). 2.3. Mating Harnessed, sexually experienced (i.e. at least 2 years mating experience), entire rams were introduced for mating on Day 50 of the experiment at a ratio of 1 ram (Suffolk and Texel) to approximately 25 ewes. The groups were not mixed prior to mating to avoid any potential effect that the groups may have on each other. Instead, the serving capacity of the entire rams was assessed (Perkins et al., 1992) and rams with medium or high serving capacity were balanced across the groups. Raddle marks were recorded daily by a single observer to identify the timing and numbers of ewes mated during the first 17 days and then recorded weekly for the subsequent 34 days. Raddle colour was changed on Day 14 after entire ram introduction to permit identification of ewes not conceiving to the first service. Ewes were maintained in accordance with commercial farm practice until lambing. During the lambing period, ewes were checked daily and the date of lambing recorded.
252
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
2.4. Blood collection and hormone analysis Blood was sampled every 3–4 days from a subset of SVR (n = 22), RVR (n = 22), PVR (n = 22) and NVR ewes (n = 22) for the duration of the pre-mating period. Progesterone profiles were produced for each animal to establish if they were anovulatory prior to the first exposure period and to monitor the distribution of oestrus over the pre-mating period. Blood samples were centrifuged within 2 h of collection at 2000 g for 20 min. Plasma was decanted into duplicate plastic tubes that were capped, immediately frozen and stored at −20 ◦ C until analysis. Plasma progesterone concentrations were analysed in duplicate using a commercial enzymelinked immunoassay (ELISA) kit (Ridgeway Science Ltd., Gloucester, UK) according to the manufacturer instructions outlined in detail previously (Hawken et al., 2007a). The limit of sensitivity of the assay was 0.2 ng/mL. For low (1.46 ng/mL), medium (2.31 ng/mL) and high (7.00 ng/mL) concentration samples were used to estimate mean intra-assay coefficients of variation (7.5%, 5.7% and 5.2%) and mean inter-assay coefficients of variation (10.3%, 14.6% and 12.8%.). 2.5. Data analysis The normality of each data set was assessed using the Anderson Darling test for normality (Minitab 13.1, Minitab Ltd., Coventry, UK). As the data were deemed to differ significantly from a normal distribution (P < 0.05), non-parametric analyses were used throughout. 2.6. Mating The median time of mating and lambing were initially compared by Kruskal–Wallis test (Minitab 13.1) to detect any difference between the four treatments. Where a significant difference was detected, treatment groups were compared by Mann–Whitney U-test (Minitab 13.1) to detect the origin of the difference. Levene’s test (Minitab 13.1) was used to assess the homogeneity of variance around the median time of mating and lambing. The total numbers of ewes mated and lambing (expressed as a % of ewes mated each day after entire ram introduction or lambed each day after the onset of lambing) were analysed using Chi-square analysis (Minitab 13.1). 2.7. Pre-mating: raddle mark data The numbers of ewes marked by the vasectomised rams in PVR and NVR ewes were recorded twice weekly to monitor the timing and distribution of oestrus during the pre-mating period. The number of days from the date of novel ram exposure or raddle colour change to marking were compared within PVR and NVR ewes using Wilcoxon Signed Rank test (Genstat 5 for Windows, Second Edition, Lawes Agricultural Trust, Rothamsted Experimental Station, UK) and between groups using Mann–Whitney U-test (Minitab 13.1). Similar analysis was not possible in SVR and RVR ewes as the rams were removed after and between exposure periods. 2.8. Pre-mating: blood sampled ewes Ewes were initially assessed for their anovulatory or cyclic status, as for the purpose of this study we only wanted to consider ewes that were anovulatory at the time of the first vasectomised
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
253
ram exposure. A ewe was classified as cyclic and thus excluded from analysis if progesterone concentrations were at or above 1.5 ng/mL for at least two consecutive samples prior to the first ram exposure period. Among the treatments, 2 SVR, 2 RVR, 3 PVR and 1 NVR ewes were cyclic prior to the first exposure period and were thus excluded from further analysis. To establish if there were any changes in the distribution of oestrous cycles during the premating period, the onset and length of each oestrous cycle was characterised. The onset of dioestrus of each cycle was defined as the date when progesterone concentrations were at or above 1.5 ng/mL for at least two consecutive samples. Absolute cycle length was calculated as the number of days between the onset of dioestrus of two consecutive cycles. The oestrous cycles were described relative to mating, so that the three cycles prior to mating were termed cycle −3, cycle −2 and cycle −1. However, as blood sampling did not continue into the mating period, the length of the cycle immediately prior to mating (cycle −1) was calculated using adjusted raddle mark data. Specifically the mean number of days from the nadir point on the progesterone profile to the calculated date of dioestrus in previous cycles was added to the date of marking by the ram (4.72, 4.52, 4.11, 5.06 days for SVR, RVR, PVR, NVR ewes, respectively). The onset of cyclic activity and absolute cycle length were initially compared between treatments using Kruskal–Wallis test (Minitab 13.1). Where a significant difference was detected, each treatment group was compared using the Mann–Whitney U-test (Minitab 13.1) to detect the origin of the difference. Levene’s test (Minitab 13.1) was used to assess the homogeneity of variance around the median cycle length between treatments over the three cycles prior to mating. The synchrony of dioestrus at mating was calculated using the ‘last months method’ used to analyse menstrual synchrony (Weller and Weller, 1995, 1997) and validated for use in sheep elsewhere (Hawken et al., 2007a). In brief, a synchrony score is calculated for each ewe as the absolute difference in the date of dioestrous onset between that ewe and every other ewe within the group. The mean of each ewe’s synchrony score is the ‘group’ synchrony score with a low score indicating a high degree of synchrony or compaction of mating. The group synchrony scores were initially compared between the four treatment groups using the Kruskal–Wallis test (Minitab 13.1) in accordance with the method outlined by Weller and Weller (1995, 1997). Where a significant difference was detected, the group synchrony scores of each group were compared using the Mann–Whitney U-test (Minitab 13.1). Differences in the synchrony score between groups indicate variation in the distribution rather than timing of mating. 3. Results 3.1. Pre-mating data: raddle mark data (all ewes) The vasectomised rams did not mark any SVR ewes during the 24 h ram exposure. Among the RVR ewes, zero ewes were marked after the first 24 h exposure period, 4% (4/113) of ewes were marked after the second 24 h exposure and 15% (17/113) were marked after the third 24 h exposure. Among the PVR ewes, 3% (3/104) were marked between Days 0 and 17, 89% (93/104) were marked between Days 17 and 34 and 96% (100/104) were marked between Days 34 and 50. Among the NVR ewes, 3% (3/113) were marked between Days 0 and 17, 84% (95/113) were marked between Days 17 and 34 and 93% (105/113) were marked between Days 35 and 50. Fig. 1 shows the progressive deviation in the pattern of raddle mark data among cyclic PVR and NVR ewes over the three cycles prior to mating.
254
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Fig. 1. Distribution of raddle mark data over the three cycles prior to mating and at mating in PVR (grey square) and NVR ewes (black diamond). Data are expressed as the percentage of cyclic ewes marked over each 17 days period after raddle colour change or novel ram introduction. The box below the cycle bar indicates the number of ewes classified as cyclic during each 17 days period within each treatment group. As raddle mark data were collected twice weekly during the pre-mating period and daily during the mating period the mating data were adjusted to every 3–4 days to mimic the frequency of recording during the pre-mating period.
3.2. Pre-mating data: cycle data (blood sampled ewes) PVR and NVR ewes had an earlier onset of cyclic activity than RVR ewes (Table 1; P < 0.01). However, only NVR ewes differed from SVR ewes in this variable (Table 1; P < 0.05). There were no differences in the onset of cyclic activity among the continuous (PVR versus NVR) or short-term (SVR versus RVR) ram exposure groups (Table 1; P > 0.05). All sampled PVR and NVR ewes had three oestrous cycles prior to mating compared to 8/20 SVR and 5/20 RVR ewes (Table 1; P < 0.05). Among the remaining SVR ewes, 9/20 had two cycles and 3/20 had one cycle prior to mating. Among the remaining RVR ewes, 13/20 had two cycles prior to mating and 2/20 had one cycle prior to mating. There were no differences between any of the ram exposure groups in absolute cycle length or variance around median cycle length (Table 1; P > 0.05). 3.3. Mating data (all ewes) The median date of mating occurred 1 day earlier in NVR than PVR ewes (Table 2; P < 0.05). There were no other differences in this variable between ram exposure groups (Table 2; P > 0.05). On Day 1 after entire ram introduction, more SVR and RVR ewes were mated than PVR and NVR ewes (23/109 and 24/113 versus 10/104 and 11/113, respectively; P < 0.05) however, this difference was not sustained after Day 1. PVR and NVR ewes had less variance around the median time of mating than SVR ewes indicating a more synchronised distribution of mating (Fig. 2; P < 0.001). NVR and PVR ewes tended to have less variance around the median time of mating than RVR ewes, however, this failed to reach significance (Fig. 2; P < 0.1). There were no
Number of ewes Cyclic onset Median days from date of first ram exposure to onset of breeding season (interquartile range) Corresponding date
SVR
RVR
PVR
NVR
20
20
19
21
15.0ac (8.00–33.0)
22.0a (15.0–29.0)
10.0bc (10.0–13.0)
10.0bd (10.0–11.5)
26 September
3 October
22 September
22 September
18.0 (14.0–21.0) (n = 5)
17.0 (17.0–21.0) (n = 19)
17.0 (14.0–17.0) (n = 21)
17.0 (14.0–18.0) (n = 18)
18.0 (14.0–18.0) (n = 19)
18.0 (14.0–18.0) (n = 21)
16.1 (15.1–18.1): (n = 20)
16.2 (14.7–17.5): (n = 19)
16.5 (14.5–17.5): (n = 21)
Median oestrous cycle length over the three cycles prior to mating Cycle −3 (interquartile range): ewes having three 17.0 (14.0–17.1) (n = 8) cycles prior to mating Cycle −2 (interquartile range): ewes having two 17.0 (15.5–18.0) (n = 17) or three cycles prior to mating Cycle −1 (interquartile range): ewes having one, 16.1 (15.1–18.1): (n = 20) two or three cycles prior to mating Significant differences are indicated by different superscripts (P < 0.05).
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Table 1 Effect of continuous and short-term ram exposure on oestrous cycle dynamics, calculated using individual progesterone profiles
255
256
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Table 2 Effect of continuous and short-term ram exposure on mating, lambing and fertility SVR Number of ewes Median days from entire ram introduction to mating (interquartile range) Ewes lambing to first service: (%) Median days from entire ram introduction to lambing (interquartile range) Ewes lambing to subsequent services (%) Ewes that were barren, aborted or died (%)
RVR
PVR
NVR
109 113 104 6.0ab (3.0–7.0) 6.0b (5.0–7.0) 5.0ab (2.0–8.0)
113 5.0a (3.0–7.0)
96 (88) 151a (148–154)
97 (86) 151a (150–153)
100 (88) 154b (151–156)
92 (88) 152a (151–154)
5 (5)
7 (6)
4 (4)
10 (8)
8 (7)
6 (5)
8 (8)
6 (5)
Significant differences are indicated by different superscripts (at least P < 0.05).
Fig. 2. Cumulative distribution of mating and lambing in continuous (mating 2a—black bars: NVR, n = 113; dark grey bars: PVR, n = 104; lambing 2b: NVR, n = 97; PVR, n = 92) and short-term (mating 2c—light grey bars: RVR, n = 113; open bars: SVR, n = 109; lambing 2d: RVR, n = 100, SVR, n = 92) ram exposed groups.
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
257
differences in this variable among the continuous (PVR versus NVR) or short-term (SVR versus RVR) ram exposure groups (P > 0.05). 3.4. Mating data (blood sampled ewes) PVR and NVR ewes had lower synchrony scores than SVR and RVR ewes indicating a more synchronised distribution of mating (1.06 ± 0.15 and 1.06 ± 0.10 versus 3.42 ± 0.32 and 2.64 ± 0.15; P < 0.001). RVR ewes had a lower synchrony score than SVR ewes indicating that the synchrony of mating was improved by repeating the short-term ram exposure (P < 0.05). This finding is supported by the distribution of mating of the groups as a whole (Fig. 2). There was no difference in synchrony scores among the continuous ram exposure groups (PVR: 1.06 ± 0.15 and NVR: 1.06 ± 0.10; P > 0.05). The absence of any difference in the synchrony scores of PVR and NVR ewes, in contrast to the visible difference in the distributions of mating of the group (Fig. 2), is due to the different median times of mating rather than the distribution of mating itself. This theory is supported by the visible shift in the cumulative distribution of raddle mark data over the pre-mating period in the NVR ewes, that was not evident in PVR ewes (Fig. 3a versus b). 3.5. Lambing data (all ewes) Within ewes lambing to first service, SVR, PVR, NVR ewes lambed at least 2 days earlier than RVR ewes (Table 2; P < 0.01). However, by Day 7 after the onset of lambing of each group,
Fig. 3. Cumulative distribution of raddle mark data for NVR (3a; black, closed symbols) and PVR ewes (3b; grey, open symbols) over the three oestrous cycles prior to mating (cycle −3: diamond, NVR; n = 3, PVR; n = 3; cycle −2: square, NVR; n = 95, PVR; n = 93; cycle −1: triangle, NVR; n = 105, PVR; n = 100) and at mating (dash, NVR; n = 113, PVR; n = 104). As for Fig. 1, the mating data were adjusted to every 3–4 days to mimic the frequency of recording during the pre-mating period.
258
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
more NVR and RVR ewes had lambed than PVR and SVR ewes (52/97 and 42/100 versus 22/92 and 26/96; P < 0.05). PVR and NVR ewes had less variance around the median time of lambing than SVR and RVR ewes (P < 0.05). Within the continuous ram exposure groups, NVR ewes had less variance around the median time of lambing than PVR ewes indicating a more synchronised distribution of lambing (Fig. 2; P < 0.05). There were no differences in this variable among the short-term (RVR versus SVR) ram exposure groups (Fig. 2; P > 0.05). 4. Discussion Ewes continuously exposed to rams from the transition into the core of the breeding season had an earlier onset of cyclic activity and a more synchronised distribution of mating than ewes given only short-term exposure to rams. This observation clearly supports the concept that continued ram presence is necessary to stimulate an optimal response to the ram effect (Signoret et al., 1982). However, as more repeated ram exposed ewes ovulated in response to the second ram exposure, there appears to be a positive relationship between the efficacy of a 24 h ram exposure and the proximity of the natural breeding season. This theory is supported by the direct correlation between the proportion of ewes ovulating in response to the ram effect and the numbers of ewes spontaneously ovulating, a factor that increases with the proximity of the onset of the breeding season (Lindsay and Signoret, 1980). Among the continuous ram exposed ewes, replacement of rams every 17 days with ‘novel’ rams altered the distribution of mating and improved the synchrony of lambing. Though this change was relatively small, it is important to note that the two groups had a similar distribution of oestrus at the onset of cyclic activity that diverged over time in the novel ram exposed ewes. This finding challenges two important aspects of the ram effect, firstly the recommendation to isolate ewes prior to stimulation by rams and secondly the ability of cyclic ewes to respond to rams. The ability of the novel males to change the distribution of oestrus is supported by the capacity of females maintained in continuous contact with males to ovulate in response to an unfamiliar or ‘novel’ male (sheep, Pearce and Oldham, 1988; Cushwa et al., 1992; goats, Veliz et al., 2006). However, in this study, the raddle mark data indicated that 84% and 93% of the novel ram exposed ewes were cyclic when the first and second set of novel rams were introduced, indicating an effect of the novel rams on cyclic ewes. A recent study found that ewes at all stages of the oestrous cycle respond to rams with an increase in pulsatile LH and proposed that this may impact on the timing and distribution of oestrus by affecting the inter-oestrus interval (Hawken et al., 2007b). Therefore, we propose that the sequential removal of rams and replacement with novel rams elicits an endocrine response that affects the distribution of cyclic activity within the group compared with ewes maintained with the same rams. Among the short-term ram exposed ewes, the distribution of mating in the repeated ram exposed ewes was less protracted than the single ram exposed ewes with two peaks of oestrus activity on Days 1 and 6 of mating. This bimodal distribution is characteristic of the ram effect (Oldham and Martin, 1978) and is driven by a proportion of ewes having short cycles or delayed ovulation in response to rams (Ungerfeld et al., 2002). In contrast, the distribution of oestrus in the single ram exposed ewes peaked on Day 1 after mating. Though the distribution of mating as a whole was more protracted than the repeated or continuous ram exposed ewes, this peak was unexpected, as the ewes had only had a single 24-h exposure to rams, 50 days previously. Among the single ram exposed ewes sampled for progesterone, 40% ovulated in response to the 24-h ram exposure. This level of responsiveness was markedly higher than expected after only 24-h of ram exposure (25% RVR ewes, current study; 19%, Signoret et al., 1982; 21%, Hawken et al., 2007a). However,
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
259
the high proportion of ewes ovulating in response to the 24-h ram exposure may have been the driving force behind the distribution of oestrus, three cycles later at mating. The stimulatory effect of oestrous females on anovulatory females is well demonstrated in goats (Walkden-Brown et al., 1993), but remains less clear in sheep (O’Callaghan et al., 1994; Zarco et al., 1995; Yildiz et al., 2002, 2004). However, those studies that found a stimulatory effect of oestrous females used a high ratio of oestrus to anovulatory females (50%, Zarco et al., 1995). Therefore, in the absence of subsequent stimulation by rams, the high proportion of ewes initially stimulated to ovulate by rams may have driven the onset of cyclic activity of the remaining anovulatory ewes. In conclusion, ewes in continuous contact with rams during the transition into the breeding season had a more synchronised distribution of mating and lambing than ewes given only shortterm contact with rams. The distribution of mating and lambing in ewes in continuous contact with rams can be further enhanced by exposure to ‘novel’ rams every 17 days. The concepts described in this paper need to be tested in other breeds and latitudes to improve our understanding of the importance of ram novelty in the ram effect. Acknowledgements We thank D. Routledge, A. Fogerty and J. Wightman for their assistance in the care and management of the animals within this experiment and the Yorkshire Agricultural Society for their funding assistance. We also thank C. Bulman, M. Hearn, S. Madgwick and R. Pickard for their assistance in the data collection. References Cushwa, W.T., Bradford, G.E., Stabenfeldt, G.H., Berger, Y.M., Dally, M.R., 1992. Ram influence on ovarian and sexual activity in anestrous ewes: effects of isolation of ewes from rams before joining and date of ram introduction. J. Anim. Sci. 70, 1195–1200. Hawken, P.A.R., Evans, A.C.O., Beard, A.P., 2008. Short term, repeated exposure to rams during the transition into the breeding season improves the synchrony of mating in the breeding season. Anim. Reprod. Sci. 106, 333–344. Hawken, P.A.R., Beard, A.P., Esmaili, T., Kadokawa, H., Evans, A.C.O., Blache, D., Martin, G.B., 2007b. The introduction of rams induces an increase in pulsatile LH secretion in cyclic ewes during the breeding season. Theriogenology 68, 56–66. Lindsay, D.R., Signoret, J.P., 1980. Influence of behaviour on reproduction. Proc. Aust. Soc. Anim. Prod. 1, 83–92. Martin, G.B., Oldham, C.M., Cognie, Y., Pearce, D.T., 1986. The physiological response of anovulatory ewes to the introduction of rams—a review. Livestock Prod. Sci. 15, 219–247. O’Callaghan, D., Donovan, A., Sunderland, S.J., Boland, M.P., Roche, J.F., 1994. Effect of the presence of male and female flockmates on reproductive activity in ewes. J. Reprod. Fertil. 100, 497–503. Oldham, C.M., Martin, G.B., 1978. Stimulation of seasonally anovular Merino ewes by rams. II. Premature regression of ram induced corpora lutea. Anim. Reprod. Sci. 1, 291–295. Pearce, D.T., Oldham, C.M., 1988. Importance of non-olfactory stimuli in mediating ram induced ovulation in the ewe. J. Reprod. Fertil. 84, 333–339. Perkins, A., Fitzgerald, J.A., Price, E.O., 1992. Sexual performance of rams in serving capacity tests predicts success in pen breeding. J. Anim. Sci. 70, 2722–2725. Rosa, H.J., Bryant, M.J., 2002. The ram effect as a way of modifying the reproductive activity in the ewe: a review. Small Rumin. Res. 45, 1–16. Signoret, J.P., Fulkerson, W.J., Lindsay, D.R., 1982. Effectiveness of testosterone treated wethers and ewes as teasers. Appl. Anim. Ethol. 9, 37–45. Ungerfeld, R., Forsberg, M., Rubianes, E., 2004. Overview of the response of anoestrous ewes to the ram effect. Reprod. Fertil. Dev. 16 (4), 479–490.
260
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Ungerfeld, R., Pinczak, A., Forsberg, M., Rubianes, E., 2002. Ovarian responses of anoestrous ewes to the “ram effect”. Can. J. Anim. Sci. 82 (4), 599–602. Veliz, F.G., Poindron, P., Malpaux, B., Delgadillo, J.A., 2006. Maintaining contact with bucks does not induce refractoriness to the male effect in seasonally anoestrous female goats. Anim. Reprod. Sci. 92 (3–4), 300–309. Walkden-Brown, S.W., Restall, B.J., Henniawati, 1993. The male effect in the Australian cashmere goat. 3. Enhancement with buck nutrition and the use of estrous females. Anim. Reprod. Sci. 32 (1–2), 69–84. Weller, A., Weller, L., 1995. Examination of menstrual synchrony among women basketball players. Psychoneuroendocrinology 20 (6), 613–622. Weller, L., Weller, A., 1997. Menstrual variability and the measurement of menstrual synchrony. Psychoneuroendocrinology 22 (2), 115–128. Yildiz, S., Metehan, U., Mehmet, K., 2004. Effects of rams and luteal or follicular phase ewes on preovulatory LH surge characteristics in ewes. Turk. J. Vet. Anim. Sci. 28, 669–673. Yildiz, S., Uzun, M., Cenesiz, M., Ucar, O., Kaya, M., Onder, F., 2002. Effects of sexually activated rams or ewes on pulsatile LH secretion in anoestrous sheep. Acta Vet. 71 (3), 297–302. Zarco, L., Rodriguez, E.F., Angulo, M.R.B., Valencia, J., 1995. Female to female stimulation of ovarian activity in the ewe. Anim. Reprod. Sci. 39, 251–258.
Ram novelty and the duration of ram exposure affects the distribution of mating in ewes exposed to rams during the transition into the breeding season P.A.R. Hawken ∗ , A.P. Beard School of Agriculture, Food and Rural Development, Newcastle University, Newcastle Upon Tyne NE1 7RU, United Kingdom Received 4 October 2007; received in revised form 27 February 2008; accepted 14 March 2008 Available online 20 March 2008
Abstract This study compared the affect of short-term and continuous exposure to rams during the transition between anoestrus and the breeding season on the distribution of mating and subsequent lambing. Further, within ewes continuously exposed to rams we investigated the effect of replacing these rams every 17 days with ‘novel’ rams. During August (late anoestrus, Northern Hemisphere), multiparous, North of England mule ewes were allocated to one of four groups: SVR ewes were exposed to vasectomised rams for 24 h on Day 0 (short term; n = 109), RVR ewes were exposed to vasectomised rams for 24 h on Days 0, 17 and 34 (short term; n = 113); PVR ewes were exposed to vasectomised rams on Day 0 and remained with the same rams for the duration of the pre-mating period (continuous; n = 104); NVR ewes were continuously exposed to vasectomised rams from Day 0 with the rams replaced with ‘novel’ rams every 17 days (continuous; n = 113). Blood samples were collected from a subset of ewes (n = 22 per group) to monitor progesterone. On Day 50, harnessed, entire rams were introduced for mating and raddle marks recorded daily for the first 17 days. The median date of mating occurred 1 day earlier in NVR ewes than PVR ewes (P < 0.05). A synchrony score calculated from the blood sampled ewes showed that the distribution of mating was more synchronised in PVR and NVR ewes than SVR and RVR ewes (P < 0.001). PVR and NVR ewes had an earlier onset of cyclic activity than RVR ewes (P < 0.01). However, only NVR ewes differed from SVR ewes in this variable (P < 0.05). Within ewes lambing to first service, the median date of lambing of PVR, NVR and SVR ewes occurred at least 2 days earlier than RVR ewes (at least P < 0.05). Further, PVR and NVR ewes had a more compact distribution of lambing than SVR and RVR ewes (P < 0.05) and lambing was more compact in NVR ewes than PVR ewes (P < 0.05). In conclusion, ewes in continuous contact with rams
∗ Corresponding author. Present address: M085, School of Animal Biology, Faculty of Natural and Agricultural Sciences, University of Western Australia, 35 Stirling Highway, Crawley, WA 6009, Australia. Tel.: +61 8 6488 3588; fax: +61 8 6488 1029. E-mail address: [email protected] (P.A.R. Hawken).
0378-4320/$ – see front matter © 2008 Published by Elsevier B.V. doi:10.1016/j.anireprosci.2008.03.009
250
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
prior to mating had a more synchronised distribution of mating and lambing than ewes given only short-term exposure to rams. This distribution of mating in continuous ram exposed ewes can be further enhanced by periodic exposure to novel rams. © 2008 Published by Elsevier B.V. Keywords: Ram effect; Breeding season; Novelty; Anoestrus; Ewes
1. Introduction The ram effect is the physiological response of ewes to rams and is a well-established method of stimulating ovulation and oestrus in anovulatory ewes (reviews; Martin et al., 1986; Rosa and Bryant, 2002; Ungerfeld et al., 2004). However, its use in genotypes native to temperate regions is restricted by breed seasonality (Lindsay and Signoret, 1980). Within these highly seasonal breeds, the proportion of ewes ovulating in response to rams only reaches acceptable levels close to the onset of the natural breeding season (Rosa and Bryant, 2002). Therefore, though the ram effect is relatively ineffective at stimulating ovulation during anoestrus in these breeds, rams can be used to control the onset of cyclic activity and the distribution of mating during the breeding season itself (Hawken et al., 2007a). In our previous study, we found that repeated, short-term exposure of North of England mule ewes to rams during the transition into the breeding season compacted the distribution of mating compared with ewes isolated from rams prior to mating (Hawken et al., 2007a). However, this premating strategy may be improved by leaving the rams with the ewes until mating, as the ovulatory response to the ram effect is positively correlated with the duration of ram presence (Signoret et al., 1982). One potential problem with this theory is that ewes maintained in continuous ram contact can become habituated or refractory to the male stimulus (Martin et al., 1986). However, more recent studies in sheep and goats indicate that habituation to male presence can be overcome by frequent replacement of males with unfamiliar or ‘novel’ males (Cushwa et al., 1992; Veliz et al., 2006). In this study we compared the effect of continuous and short-term exposure to rams during the transition between anoestrus and the breeding season on the distribution of mating later during the breeding season and subsequent lambing. Furthermore, in ewes continuously exposed to rams, we tested if replacement of these rams with ‘novel’ rams every 17 days would affect the distribution of mating and lambing. 2. Materials and methods 2.1. Animals and experimental procedures The experiment was carried out in accordance with the Animals (Scientific Procedures) Act 1986 and was approved by the University of Newcastle Animal Ethics committee and the UK Home Office. During August (late anoestrus, Northern Hemisphere) multiparous, North of England mule ewes (Scottish Blackface × Bluefaced Leicester) that had been previously isolated from rams (not within 500 m for a minimum of 2 months) were allocated to either short-term or continuous ram exposure groups. The groups which tested the efficacy of short-term exposure to rams were
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
251
single ram exposure ewes (SVR; n = 109) and repeated ram exposure ewes (RVR; n = 113). The groups that tested the efficacy of continuous ram exposure were permanent ram exposure ewes (PVR; n = 104) and novel ram exposure ewes (NVR; n = 113). All groups were balanced for age (4–6 years) and maintained on pasture at Cockle Park Farm, Northumberland, UK (55◦ 13 N). 2.2. Pre-mating period SVR and RVR ewes were exposed to harnessed, vasectomised rams in the intermittent exposure paddock (0.6 ha) for 24 h on Day 0 of the experiment (11 September). SVR ewes were subsequently isolated from rams for the remainder of the pre-mating period in a paddock more than 500 m from the exposure paddock. RVR ewes were also exposed to harnessed, vasectomised rams for 24 h on Days 17 and 34 of the experiment. Between each 24 h exposure period, the RVR ewes were isolated from rams in a paddock more than 500 m from the exposure paddock. To provide a degree of ram novelty during each exposure, the RVR ewes had not previously been exposed to one of the rams; the other two vasectomised rams remained the same for all three exposure periods. PVR ewes were exposed to harnessed, vasectomised rams on Day 0. The ewes remained with these rams for the duration of the pre-mating period. NVR ewes were exposed to harnessed, vasectomised rams on Day 0. The ewes remained with these rams for the first 17 days until they were replaced with ‘novel’ rams on Days 17 and 34 of the experiment. All paddocks were between 4.5 and 5.0 ha and sown with a comparable pasture mix (rye grass/clover mix). No direct measurements of pasture growth or quality were taken, but visual assessment of the paddocks indicated comparable food availability between paddocks throughout the experiment. The paddocks where rams were permanently present (PVR and NVR) were 6 km from the intermittent exposure paddocks to prevent any impact of these rams on SVR and RVR ewes. The vasectomised rams were crossbred (Scottish Blackface or Suffolk × Border Leicester; n = 3 per exposure) and sexually experienced (i.e. at least 2 years mating experience). They were fitted with harnesses to allow detection of oestrus and were in direct physical contact with the ewes. Raddle marks were recorded by a single observer after each 24 h exposure period in the SVR and RVR ewes and twice weekly in the PVR and NVR ewes. The raddle colour was changed before each 24 h exposure period of RVR ewes, on Days 17 and 34 in the PVR ewes and with each set of novel rams in the NVR ewes. When the vasectomised rams were not being used to stimulate ewes, they were kept with other rams and isolated from ewes (i.e. more than 500 m). 2.3. Mating Harnessed, sexually experienced (i.e. at least 2 years mating experience), entire rams were introduced for mating on Day 50 of the experiment at a ratio of 1 ram (Suffolk and Texel) to approximately 25 ewes. The groups were not mixed prior to mating to avoid any potential effect that the groups may have on each other. Instead, the serving capacity of the entire rams was assessed (Perkins et al., 1992) and rams with medium or high serving capacity were balanced across the groups. Raddle marks were recorded daily by a single observer to identify the timing and numbers of ewes mated during the first 17 days and then recorded weekly for the subsequent 34 days. Raddle colour was changed on Day 14 after entire ram introduction to permit identification of ewes not conceiving to the first service. Ewes were maintained in accordance with commercial farm practice until lambing. During the lambing period, ewes were checked daily and the date of lambing recorded.
252
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
2.4. Blood collection and hormone analysis Blood was sampled every 3–4 days from a subset of SVR (n = 22), RVR (n = 22), PVR (n = 22) and NVR ewes (n = 22) for the duration of the pre-mating period. Progesterone profiles were produced for each animal to establish if they were anovulatory prior to the first exposure period and to monitor the distribution of oestrus over the pre-mating period. Blood samples were centrifuged within 2 h of collection at 2000 g for 20 min. Plasma was decanted into duplicate plastic tubes that were capped, immediately frozen and stored at −20 ◦ C until analysis. Plasma progesterone concentrations were analysed in duplicate using a commercial enzymelinked immunoassay (ELISA) kit (Ridgeway Science Ltd., Gloucester, UK) according to the manufacturer instructions outlined in detail previously (Hawken et al., 2007a). The limit of sensitivity of the assay was 0.2 ng/mL. For low (1.46 ng/mL), medium (2.31 ng/mL) and high (7.00 ng/mL) concentration samples were used to estimate mean intra-assay coefficients of variation (7.5%, 5.7% and 5.2%) and mean inter-assay coefficients of variation (10.3%, 14.6% and 12.8%.). 2.5. Data analysis The normality of each data set was assessed using the Anderson Darling test for normality (Minitab 13.1, Minitab Ltd., Coventry, UK). As the data were deemed to differ significantly from a normal distribution (P < 0.05), non-parametric analyses were used throughout. 2.6. Mating The median time of mating and lambing were initially compared by Kruskal–Wallis test (Minitab 13.1) to detect any difference between the four treatments. Where a significant difference was detected, treatment groups were compared by Mann–Whitney U-test (Minitab 13.1) to detect the origin of the difference. Levene’s test (Minitab 13.1) was used to assess the homogeneity of variance around the median time of mating and lambing. The total numbers of ewes mated and lambing (expressed as a % of ewes mated each day after entire ram introduction or lambed each day after the onset of lambing) were analysed using Chi-square analysis (Minitab 13.1). 2.7. Pre-mating: raddle mark data The numbers of ewes marked by the vasectomised rams in PVR and NVR ewes were recorded twice weekly to monitor the timing and distribution of oestrus during the pre-mating period. The number of days from the date of novel ram exposure or raddle colour change to marking were compared within PVR and NVR ewes using Wilcoxon Signed Rank test (Genstat 5 for Windows, Second Edition, Lawes Agricultural Trust, Rothamsted Experimental Station, UK) and between groups using Mann–Whitney U-test (Minitab 13.1). Similar analysis was not possible in SVR and RVR ewes as the rams were removed after and between exposure periods. 2.8. Pre-mating: blood sampled ewes Ewes were initially assessed for their anovulatory or cyclic status, as for the purpose of this study we only wanted to consider ewes that were anovulatory at the time of the first vasectomised
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
253
ram exposure. A ewe was classified as cyclic and thus excluded from analysis if progesterone concentrations were at or above 1.5 ng/mL for at least two consecutive samples prior to the first ram exposure period. Among the treatments, 2 SVR, 2 RVR, 3 PVR and 1 NVR ewes were cyclic prior to the first exposure period and were thus excluded from further analysis. To establish if there were any changes in the distribution of oestrous cycles during the premating period, the onset and length of each oestrous cycle was characterised. The onset of dioestrus of each cycle was defined as the date when progesterone concentrations were at or above 1.5 ng/mL for at least two consecutive samples. Absolute cycle length was calculated as the number of days between the onset of dioestrus of two consecutive cycles. The oestrous cycles were described relative to mating, so that the three cycles prior to mating were termed cycle −3, cycle −2 and cycle −1. However, as blood sampling did not continue into the mating period, the length of the cycle immediately prior to mating (cycle −1) was calculated using adjusted raddle mark data. Specifically the mean number of days from the nadir point on the progesterone profile to the calculated date of dioestrus in previous cycles was added to the date of marking by the ram (4.72, 4.52, 4.11, 5.06 days for SVR, RVR, PVR, NVR ewes, respectively). The onset of cyclic activity and absolute cycle length were initially compared between treatments using Kruskal–Wallis test (Minitab 13.1). Where a significant difference was detected, each treatment group was compared using the Mann–Whitney U-test (Minitab 13.1) to detect the origin of the difference. Levene’s test (Minitab 13.1) was used to assess the homogeneity of variance around the median cycle length between treatments over the three cycles prior to mating. The synchrony of dioestrus at mating was calculated using the ‘last months method’ used to analyse menstrual synchrony (Weller and Weller, 1995, 1997) and validated for use in sheep elsewhere (Hawken et al., 2007a). In brief, a synchrony score is calculated for each ewe as the absolute difference in the date of dioestrous onset between that ewe and every other ewe within the group. The mean of each ewe’s synchrony score is the ‘group’ synchrony score with a low score indicating a high degree of synchrony or compaction of mating. The group synchrony scores were initially compared between the four treatment groups using the Kruskal–Wallis test (Minitab 13.1) in accordance with the method outlined by Weller and Weller (1995, 1997). Where a significant difference was detected, the group synchrony scores of each group were compared using the Mann–Whitney U-test (Minitab 13.1). Differences in the synchrony score between groups indicate variation in the distribution rather than timing of mating. 3. Results 3.1. Pre-mating data: raddle mark data (all ewes) The vasectomised rams did not mark any SVR ewes during the 24 h ram exposure. Among the RVR ewes, zero ewes were marked after the first 24 h exposure period, 4% (4/113) of ewes were marked after the second 24 h exposure and 15% (17/113) were marked after the third 24 h exposure. Among the PVR ewes, 3% (3/104) were marked between Days 0 and 17, 89% (93/104) were marked between Days 17 and 34 and 96% (100/104) were marked between Days 34 and 50. Among the NVR ewes, 3% (3/113) were marked between Days 0 and 17, 84% (95/113) were marked between Days 17 and 34 and 93% (105/113) were marked between Days 35 and 50. Fig. 1 shows the progressive deviation in the pattern of raddle mark data among cyclic PVR and NVR ewes over the three cycles prior to mating.
254
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Fig. 1. Distribution of raddle mark data over the three cycles prior to mating and at mating in PVR (grey square) and NVR ewes (black diamond). Data are expressed as the percentage of cyclic ewes marked over each 17 days period after raddle colour change or novel ram introduction. The box below the cycle bar indicates the number of ewes classified as cyclic during each 17 days period within each treatment group. As raddle mark data were collected twice weekly during the pre-mating period and daily during the mating period the mating data were adjusted to every 3–4 days to mimic the frequency of recording during the pre-mating period.
3.2. Pre-mating data: cycle data (blood sampled ewes) PVR and NVR ewes had an earlier onset of cyclic activity than RVR ewes (Table 1; P < 0.01). However, only NVR ewes differed from SVR ewes in this variable (Table 1; P < 0.05). There were no differences in the onset of cyclic activity among the continuous (PVR versus NVR) or short-term (SVR versus RVR) ram exposure groups (Table 1; P > 0.05). All sampled PVR and NVR ewes had three oestrous cycles prior to mating compared to 8/20 SVR and 5/20 RVR ewes (Table 1; P < 0.05). Among the remaining SVR ewes, 9/20 had two cycles and 3/20 had one cycle prior to mating. Among the remaining RVR ewes, 13/20 had two cycles prior to mating and 2/20 had one cycle prior to mating. There were no differences between any of the ram exposure groups in absolute cycle length or variance around median cycle length (Table 1; P > 0.05). 3.3. Mating data (all ewes) The median date of mating occurred 1 day earlier in NVR than PVR ewes (Table 2; P < 0.05). There were no other differences in this variable between ram exposure groups (Table 2; P > 0.05). On Day 1 after entire ram introduction, more SVR and RVR ewes were mated than PVR and NVR ewes (23/109 and 24/113 versus 10/104 and 11/113, respectively; P < 0.05) however, this difference was not sustained after Day 1. PVR and NVR ewes had less variance around the median time of mating than SVR ewes indicating a more synchronised distribution of mating (Fig. 2; P < 0.001). NVR and PVR ewes tended to have less variance around the median time of mating than RVR ewes, however, this failed to reach significance (Fig. 2; P < 0.1). There were no
Number of ewes Cyclic onset Median days from date of first ram exposure to onset of breeding season (interquartile range) Corresponding date
SVR
RVR
PVR
NVR
20
20
19
21
15.0ac (8.00–33.0)
22.0a (15.0–29.0)
10.0bc (10.0–13.0)
10.0bd (10.0–11.5)
26 September
3 October
22 September
22 September
18.0 (14.0–21.0) (n = 5)
17.0 (17.0–21.0) (n = 19)
17.0 (14.0–17.0) (n = 21)
17.0 (14.0–18.0) (n = 18)
18.0 (14.0–18.0) (n = 19)
18.0 (14.0–18.0) (n = 21)
16.1 (15.1–18.1): (n = 20)
16.2 (14.7–17.5): (n = 19)
16.5 (14.5–17.5): (n = 21)
Median oestrous cycle length over the three cycles prior to mating Cycle −3 (interquartile range): ewes having three 17.0 (14.0–17.1) (n = 8) cycles prior to mating Cycle −2 (interquartile range): ewes having two 17.0 (15.5–18.0) (n = 17) or three cycles prior to mating Cycle −1 (interquartile range): ewes having one, 16.1 (15.1–18.1): (n = 20) two or three cycles prior to mating Significant differences are indicated by different superscripts (P < 0.05).
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Table 1 Effect of continuous and short-term ram exposure on oestrous cycle dynamics, calculated using individual progesterone profiles
255
256
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Table 2 Effect of continuous and short-term ram exposure on mating, lambing and fertility SVR Number of ewes Median days from entire ram introduction to mating (interquartile range) Ewes lambing to first service: (%) Median days from entire ram introduction to lambing (interquartile range) Ewes lambing to subsequent services (%) Ewes that were barren, aborted or died (%)
RVR
PVR
NVR
109 113 104 6.0ab (3.0–7.0) 6.0b (5.0–7.0) 5.0ab (2.0–8.0)
113 5.0a (3.0–7.0)
96 (88) 151a (148–154)
97 (86) 151a (150–153)
100 (88) 154b (151–156)
92 (88) 152a (151–154)
5 (5)
7 (6)
4 (4)
10 (8)
8 (7)
6 (5)
8 (8)
6 (5)
Significant differences are indicated by different superscripts (at least P < 0.05).
Fig. 2. Cumulative distribution of mating and lambing in continuous (mating 2a—black bars: NVR, n = 113; dark grey bars: PVR, n = 104; lambing 2b: NVR, n = 97; PVR, n = 92) and short-term (mating 2c—light grey bars: RVR, n = 113; open bars: SVR, n = 109; lambing 2d: RVR, n = 100, SVR, n = 92) ram exposed groups.
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
257
differences in this variable among the continuous (PVR versus NVR) or short-term (SVR versus RVR) ram exposure groups (P > 0.05). 3.4. Mating data (blood sampled ewes) PVR and NVR ewes had lower synchrony scores than SVR and RVR ewes indicating a more synchronised distribution of mating (1.06 ± 0.15 and 1.06 ± 0.10 versus 3.42 ± 0.32 and 2.64 ± 0.15; P < 0.001). RVR ewes had a lower synchrony score than SVR ewes indicating that the synchrony of mating was improved by repeating the short-term ram exposure (P < 0.05). This finding is supported by the distribution of mating of the groups as a whole (Fig. 2). There was no difference in synchrony scores among the continuous ram exposure groups (PVR: 1.06 ± 0.15 and NVR: 1.06 ± 0.10; P > 0.05). The absence of any difference in the synchrony scores of PVR and NVR ewes, in contrast to the visible difference in the distributions of mating of the group (Fig. 2), is due to the different median times of mating rather than the distribution of mating itself. This theory is supported by the visible shift in the cumulative distribution of raddle mark data over the pre-mating period in the NVR ewes, that was not evident in PVR ewes (Fig. 3a versus b). 3.5. Lambing data (all ewes) Within ewes lambing to first service, SVR, PVR, NVR ewes lambed at least 2 days earlier than RVR ewes (Table 2; P < 0.01). However, by Day 7 after the onset of lambing of each group,
Fig. 3. Cumulative distribution of raddle mark data for NVR (3a; black, closed symbols) and PVR ewes (3b; grey, open symbols) over the three oestrous cycles prior to mating (cycle −3: diamond, NVR; n = 3, PVR; n = 3; cycle −2: square, NVR; n = 95, PVR; n = 93; cycle −1: triangle, NVR; n = 105, PVR; n = 100) and at mating (dash, NVR; n = 113, PVR; n = 104). As for Fig. 1, the mating data were adjusted to every 3–4 days to mimic the frequency of recording during the pre-mating period.
258
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
more NVR and RVR ewes had lambed than PVR and SVR ewes (52/97 and 42/100 versus 22/92 and 26/96; P < 0.05). PVR and NVR ewes had less variance around the median time of lambing than SVR and RVR ewes (P < 0.05). Within the continuous ram exposure groups, NVR ewes had less variance around the median time of lambing than PVR ewes indicating a more synchronised distribution of lambing (Fig. 2; P < 0.05). There were no differences in this variable among the short-term (RVR versus SVR) ram exposure groups (Fig. 2; P > 0.05). 4. Discussion Ewes continuously exposed to rams from the transition into the core of the breeding season had an earlier onset of cyclic activity and a more synchronised distribution of mating than ewes given only short-term exposure to rams. This observation clearly supports the concept that continued ram presence is necessary to stimulate an optimal response to the ram effect (Signoret et al., 1982). However, as more repeated ram exposed ewes ovulated in response to the second ram exposure, there appears to be a positive relationship between the efficacy of a 24 h ram exposure and the proximity of the natural breeding season. This theory is supported by the direct correlation between the proportion of ewes ovulating in response to the ram effect and the numbers of ewes spontaneously ovulating, a factor that increases with the proximity of the onset of the breeding season (Lindsay and Signoret, 1980). Among the continuous ram exposed ewes, replacement of rams every 17 days with ‘novel’ rams altered the distribution of mating and improved the synchrony of lambing. Though this change was relatively small, it is important to note that the two groups had a similar distribution of oestrus at the onset of cyclic activity that diverged over time in the novel ram exposed ewes. This finding challenges two important aspects of the ram effect, firstly the recommendation to isolate ewes prior to stimulation by rams and secondly the ability of cyclic ewes to respond to rams. The ability of the novel males to change the distribution of oestrus is supported by the capacity of females maintained in continuous contact with males to ovulate in response to an unfamiliar or ‘novel’ male (sheep, Pearce and Oldham, 1988; Cushwa et al., 1992; goats, Veliz et al., 2006). However, in this study, the raddle mark data indicated that 84% and 93% of the novel ram exposed ewes were cyclic when the first and second set of novel rams were introduced, indicating an effect of the novel rams on cyclic ewes. A recent study found that ewes at all stages of the oestrous cycle respond to rams with an increase in pulsatile LH and proposed that this may impact on the timing and distribution of oestrus by affecting the inter-oestrus interval (Hawken et al., 2007b). Therefore, we propose that the sequential removal of rams and replacement with novel rams elicits an endocrine response that affects the distribution of cyclic activity within the group compared with ewes maintained with the same rams. Among the short-term ram exposed ewes, the distribution of mating in the repeated ram exposed ewes was less protracted than the single ram exposed ewes with two peaks of oestrus activity on Days 1 and 6 of mating. This bimodal distribution is characteristic of the ram effect (Oldham and Martin, 1978) and is driven by a proportion of ewes having short cycles or delayed ovulation in response to rams (Ungerfeld et al., 2002). In contrast, the distribution of oestrus in the single ram exposed ewes peaked on Day 1 after mating. Though the distribution of mating as a whole was more protracted than the repeated or continuous ram exposed ewes, this peak was unexpected, as the ewes had only had a single 24-h exposure to rams, 50 days previously. Among the single ram exposed ewes sampled for progesterone, 40% ovulated in response to the 24-h ram exposure. This level of responsiveness was markedly higher than expected after only 24-h of ram exposure (25% RVR ewes, current study; 19%, Signoret et al., 1982; 21%, Hawken et al., 2007a). However,
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
259
the high proportion of ewes ovulating in response to the 24-h ram exposure may have been the driving force behind the distribution of oestrus, three cycles later at mating. The stimulatory effect of oestrous females on anovulatory females is well demonstrated in goats (Walkden-Brown et al., 1993), but remains less clear in sheep (O’Callaghan et al., 1994; Zarco et al., 1995; Yildiz et al., 2002, 2004). However, those studies that found a stimulatory effect of oestrous females used a high ratio of oestrus to anovulatory females (50%, Zarco et al., 1995). Therefore, in the absence of subsequent stimulation by rams, the high proportion of ewes initially stimulated to ovulate by rams may have driven the onset of cyclic activity of the remaining anovulatory ewes. In conclusion, ewes in continuous contact with rams during the transition into the breeding season had a more synchronised distribution of mating and lambing than ewes given only shortterm contact with rams. The distribution of mating and lambing in ewes in continuous contact with rams can be further enhanced by exposure to ‘novel’ rams every 17 days. The concepts described in this paper need to be tested in other breeds and latitudes to improve our understanding of the importance of ram novelty in the ram effect. Acknowledgements We thank D. Routledge, A. Fogerty and J. Wightman for their assistance in the care and management of the animals within this experiment and the Yorkshire Agricultural Society for their funding assistance. We also thank C. Bulman, M. Hearn, S. Madgwick and R. Pickard for their assistance in the data collection. References Cushwa, W.T., Bradford, G.E., Stabenfeldt, G.H., Berger, Y.M., Dally, M.R., 1992. Ram influence on ovarian and sexual activity in anestrous ewes: effects of isolation of ewes from rams before joining and date of ram introduction. J. Anim. Sci. 70, 1195–1200. Hawken, P.A.R., Evans, A.C.O., Beard, A.P., 2008. Short term, repeated exposure to rams during the transition into the breeding season improves the synchrony of mating in the breeding season. Anim. Reprod. Sci. 106, 333–344. Hawken, P.A.R., Beard, A.P., Esmaili, T., Kadokawa, H., Evans, A.C.O., Blache, D., Martin, G.B., 2007b. The introduction of rams induces an increase in pulsatile LH secretion in cyclic ewes during the breeding season. Theriogenology 68, 56–66. Lindsay, D.R., Signoret, J.P., 1980. Influence of behaviour on reproduction. Proc. Aust. Soc. Anim. Prod. 1, 83–92. Martin, G.B., Oldham, C.M., Cognie, Y., Pearce, D.T., 1986. The physiological response of anovulatory ewes to the introduction of rams—a review. Livestock Prod. Sci. 15, 219–247. O’Callaghan, D., Donovan, A., Sunderland, S.J., Boland, M.P., Roche, J.F., 1994. Effect of the presence of male and female flockmates on reproductive activity in ewes. J. Reprod. Fertil. 100, 497–503. Oldham, C.M., Martin, G.B., 1978. Stimulation of seasonally anovular Merino ewes by rams. II. Premature regression of ram induced corpora lutea. Anim. Reprod. Sci. 1, 291–295. Pearce, D.T., Oldham, C.M., 1988. Importance of non-olfactory stimuli in mediating ram induced ovulation in the ewe. J. Reprod. Fertil. 84, 333–339. Perkins, A., Fitzgerald, J.A., Price, E.O., 1992. Sexual performance of rams in serving capacity tests predicts success in pen breeding. J. Anim. Sci. 70, 2722–2725. Rosa, H.J., Bryant, M.J., 2002. The ram effect as a way of modifying the reproductive activity in the ewe: a review. Small Rumin. Res. 45, 1–16. Signoret, J.P., Fulkerson, W.J., Lindsay, D.R., 1982. Effectiveness of testosterone treated wethers and ewes as teasers. Appl. Anim. Ethol. 9, 37–45. Ungerfeld, R., Forsberg, M., Rubianes, E., 2004. Overview of the response of anoestrous ewes to the ram effect. Reprod. Fertil. Dev. 16 (4), 479–490.
260
P.A.R. Hawken, A.P. Beard / Animal Reproduction Science 111 (2009) 249–260
Ungerfeld, R., Pinczak, A., Forsberg, M., Rubianes, E., 2002. Ovarian responses of anoestrous ewes to the “ram effect”. Can. J. Anim. Sci. 82 (4), 599–602. Veliz, F.G., Poindron, P., Malpaux, B., Delgadillo, J.A., 2006. Maintaining contact with bucks does not induce refractoriness to the male effect in seasonally anoestrous female goats. Anim. Reprod. Sci. 92 (3–4), 300–309. Walkden-Brown, S.W., Restall, B.J., Henniawati, 1993. The male effect in the Australian cashmere goat. 3. Enhancement with buck nutrition and the use of estrous females. Anim. Reprod. Sci. 32 (1–2), 69–84. Weller, A., Weller, L., 1995. Examination of menstrual synchrony among women basketball players. Psychoneuroendocrinology 20 (6), 613–622. Weller, L., Weller, A., 1997. Menstrual variability and the measurement of menstrual synchrony. Psychoneuroendocrinology 22 (2), 115–128. Yildiz, S., Metehan, U., Mehmet, K., 2004. Effects of rams and luteal or follicular phase ewes on preovulatory LH surge characteristics in ewes. Turk. J. Vet. Anim. Sci. 28, 669–673. Yildiz, S., Uzun, M., Cenesiz, M., Ucar, O., Kaya, M., Onder, F., 2002. Effects of sexually activated rams or ewes on pulsatile LH secretion in anoestrous sheep. Acta Vet. 71 (3), 297–302. Zarco, L., Rodriguez, E.F., Angulo, M.R.B., Valencia, J., 1995. Female to female stimulation of ovarian activity in the ewe. Anim. Reprod. Sci. 39, 251–258.