- Email: [email protected]
Factors influencing the success of transcervical insemination in Merino ewes
FACTORS INFLUENCING THE SUCCESS OF TRANSCERVICAL INSEMINATION IN MERINO EWES D.P. Windsor Sheep Industries Branch Great SolXhern Agricultural Research Institute PO Box 757 Katanning Western Australia 6317 Received for publication: Accepted:
July
20,
November
1994 10,
1994
ABSTRACT The experiments described examined the effects of a number of factors on the level of uterine insemination achieved in Merino ewes by a transcervical insemination technique (Guelph system for transcervical artificial insemination; GST-AI). Cervical penetration rate is an important limitation to the use of such methods in Merinos. Simulated insemination was performed to estimate the proportion of ewes in which a pipette could be passed through the cervix to the uterus. In Experiment 1, cervical penetration rate (n=14 to 30)was unaffected by an increase in postpartum interval at AI from 12 to 26 wk. The results of cervical penetration for individual ewes weir: found to be repeatable (PcO.05). Experiment 2 (197 ewes) revealed a clear effect of ewe parity on penetration rates in hormonally synchronized ewes during the nonbreeding season (P
Acknowledgements The Author wishes to thank T. Gardner, A. Edward and T. Ladyman for technical assistance. Thanks also to I. Dhaliwal for statistical advice and assistance, and to DI K.P. Croker for comments on the manuscript. Theriogenology 43:i 009-l 018,1995 0 1995 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0093-691 x/95/$1 0.00 SSDI 0093-691X(95)00065-G
1010
Theriogenology INTRODUCTION
The major factor limiting the fertility of ewes inseminated with frozen ram semen is the inability of ram spermatozoa to cross the cervix of the ewe after being subjected to the freezing and thawing processes (16). The exact nature of the cellular changes causing this problem are as yet undefined. Unlike fresh semen, which provides acceptable conception rates at cervical or vaginal insemination (19), frozen semen must be deposited within the uterus to ensure adequate fertility (22). The most widely practised form of uterine insemination in the ewe is laparoscopic AI (15). This procedure provides acceptable conception rates using both fresh and frozen semen (2,15,18,19) and is widely used within the Australian stud Merino industry (7). While this procedure is effective, its wider application is limited by its expense. An alternative to laparoscopic AI is the passage of a pipette into the uterus from die vagina via the cervix. This procedure is known as transcervical AI and, in its most familiar form, is widely practiced in cattle. Transcervical procedures are less capital intensive than laparoscopy, and do not involve surgical entry to the body cavity. The anaesthesia and veterinary skills associated with laparoscopy are not required. There have been a number of reports of successful transcervical AI in sheep (1,8-13,), but these techniques have not been widely adopted in Australia (6). One such transcervical technique was recently evaluated for use with Australian Merinos (23). That study indicated that the Guelph system for transcervical AI (GST-AI) could produce similar conception rates to laparoscopic AI, but only in those ewes in which semen was deposited within the uterus rather than the cervix (about 50% of ewes in the hands of Australian operators, 23). Consequently, cervical penetration rate (the proportion of ewes in which the AI pipette can be passed through the cervix) was identified as a major factor limiting the utility of this technique under Australian conditions. Cervical penetration rates are likely to be influenced by both operator skill and variation between ewes (reviewed by Eppleston and Maxwell; 6). One strategy to increase penetration rate may be to select only certain suitable classes of ewes for mating by transcervical insemination. Alternatively, it may be possible to manipulate pre insemination management to increase penetration rates. There is little information available on the ewe or management factors which influence cervical penetration rates by GST-AI in Merinos. This paper describes a series of experiments examining some of the factors which have the potential to influence the cervical penetration rate achieved in Merino ewes using the instruments and procedures developed for GST-AI. The data will be useful in designing experiments to test the effects of such factors on ewe fertility after GST-AI. The factors considered included postpartum interval at insemination, individual ewe variation, ewe parity, estrus synchronization technique and the stage of the breeding season.
1011
Theriogenology MATERIALS AND METHODS
Experiments 1 (60 ewes) and 2 (197 ewes) were conducted outside the normal breeding season for Merino ewes in Western Australia (August-December). Experiment 3 (161 ewes) was conducted within the breeding season (April-May). All insemination attempts were made by the same operator, who was experienced in the assessment procedure (23). All ewes were managed according to Western Australian commercial practices. Experiment 1 Mature (5 yr) Merino ewes were checked for lactational status 12 wk after the start of lambing. Ewes which had lambed were randomly allocated to two groups of 30 ewes each. The estrous cycles of the ewes were synchronized by insertion of progestagen sponges= for 12 to 14 d and the intramuscular injection of 300 IU PMSGb at the time of sponge withdrawal. The ewes were divided into two groups to avoid the need for almost continuous progestagen treatment which would have arisen had a single group been examined at three-weekly intervals. Transcervical insemination was attempted as described by Halbert et al (12) between 48 and 60 h after sponge withdrawal Ewes were restrained in dorsal recumbancy, and a tubular speculum (illuminated by a fibre optic light source) inserted into the vagina. Bozeman forceps were used to grasp and stab&e the cervix during the penetration attempt. A fine stainless steel pipette was introduced into the cervical OS,and gentle probing used to locate the passage through the cervix to the uterus. Each cervical penetration attempt was judged by the operator to have either succeeded or failed in passing the AI pipette through the cervix into the uterus (3). Cervical penetration was considered to have been successful when the AI pipette was passed around ah obstructions encountered and inserted into the cervix to its full length (8.5 cm). Because cervical penetration was the end point for this study, no semen was deposited in the reproductive tract. Cervical penetration attempts were made in Group 1 ewes at 14,20 and 26 wk after the commencement of lambing. Cervical penetration attempts were made in Group 2 ewes at 17 and 23 wk after the start of lambing. At the first time point for each group (14 and 17 wk for Groups 1 and 2, respectively), cervical penetration was attempted 3 times for each ewe to examine the repeatability of penetration (r&O). Ewes were released from the cradle and returned to their group between attempts. Ewes were classified into 8 classes (see Table 1) based on the success or failure of each penetration attempt. Fewer Group 1 ewes were available later in the experiment due to misidentification of animals following shearing.
e3Omg FGA, Ovakron, ESP, Johannesburg South Africa. bPregnecol, Horizon, North Ryde NSW.
1012
Theriogenology
Experiment 2 Ewes from a flock for which individual lambing histories were known were allocated to experimental groups on the basis of having lambed once (n=76), twice (t&8) or three times (n=53). Only ewes which had lambed at each opportunity were used to avoid any interactions between ewe parity and age. Postpartum interval for these ewes was about 8 mo. Estrus synchronization and simulated transcervical insemination were performed as in Experiment 1. Cervical penetration was assessed once for each ewe. Experiment 3 Cervical penetration in mature ewes (largely from the same flock as the ewes used in Experiment 2) with 1 or 2 lambings and a postpartum interval of about 13 mo, was assessed in ewes, with estrus controlled as follows: 1) Estrus was synchronized using progestagen sponges and PMSG as described for Experiment 1. Cervical passage was attempted 51 to 55 h after sponge withdrawal (n=51). 2) Estrus synchronized by two treatments (intramuscular injections 13 d apart) with synthetic prostaglandin F2_ (125pg cloprostenol)c and 300 IU PMSG at the second treatment. Cervical passage was attempted 51 to 55 hr after the second prostaglandin treatment (n=50). 3) Natural estrus was detected by harnessed vasectomized rams. Ewes were checked for raddle marks twice daily to detect the onset of estrus. Cervical passage was attempted 8 to 16 hr after fit estrus detection for each ewe (n=60). Statistical Analyses The proportions of ewes in which cervical penetration was successful for each treatment were compared by logistic regression using Genstat 5 (20). Separate models were fitted to examine the effects of postpartum interval, ewe parity, estrus synchronization treatment and stage of the breeding season on cervical penetration, and to assess the repeatability of cervical penetration. RESULTS Postpartum Interval The cervical penetration rate (expressed as ewes penetrated as a proportion of ewes attempted) for the first penetration attempt at each time point in Experiment 1 is shown in Figure 1. Logistic regression did not indicate any significant difference between penetration rates at the different times after the start of lambing (P=O.7). Individual Ewe Variation The number of ewes falling into each penetration class is shown in Table 1. More than 75% of ewes produced the same penetration result (1 signifying success and 0 failure) at each of 3 CEstrumate, Pitman Moore, North Ryde NSW.
1013
Theriogenology
attempts. Logistic regression showed no interaction between individual ewe and the result at each penetration attempt, indicating that the first result obtained for any ewe is repeatable (P
20
17
23
26
29
Postpartum interval (weeks)
Figure 1. Cervical penetration rates in 5 yr Merino ewes with increasing postpartum interval. Estrus was synchronized using sponges and PMSG during the non-breeding season. Postpartum interval did not affect penetration rate.
Table 1. Repeatability of cervical penetration in progestagen synchronized Merino ewes during the nonbreeding season. Penetration class represents the combination of success (1) or failure (0) at each of 3 attempts (n=60). Penetration at each of 3 attempts 000 001 010 011 100 101 110 111 Total
Number of ewes
%
12 0 4 1 3 2 4 34 60
20.0 0.0 6.7 1.7 5.0 3.3 6.7 56.6 100.0
1014
Theriogenology
Ewe Parity The cervical penetration rate in ewes having lambed once, twice or three times (Experiment 2) is shown in Figure 2. The rate of cervical passage improved with incmasing parity (P~0.05).
1
2
3
Ewe parity ( number of lambings)
Figure 2. The effect of ewe parity on cervical penetration rate in Merino ewes synchronized with sponges and PMSG in the non-breeding season. Penetration rates differed with ewe parity (P
1015
Theriogenology
Pr;g;;;gn
Pr”+“tafan~in Natural Estrus
Estrus synchronization treatment
Figure 3. Cervical penetration rates in Merino ewes synchronized using progestagen or prostaglandin, or at natural estrus during the breeding season. Penetration rates did not differ between treatments (P=O.O8).
(10/12)
Ewe age 3 years 0
4 years (28168)
(16176)
Breeding
Season
NOIItIreedlng
season
Figure 4. Cervical penetration rates in 3- and 4-year old ewes from Experiment 2 (nonbreeding season), and hormonally treated ewes originating from the same flock and used in Experiment 3 (breeding season). Penetration rate differed with season (PcO.05).
1016
Theriogenology DISCUSSION
Previous research has identified 2 major factors limiting the fertility of Merino ewes that are inseminated by GST-AI (23). These were failure to penetrate the cervix and perform a uterine insemination iu a large proportion of ewes, and the inability to consistently produce equivalent fertility to laparoscopic AI after successful transcervical AI. Similar results have been mported for other ewe breeds (3), and the goat doe (21). This present study considered only the first of these 2 problems. If poor cervical penetration rates cannot be overcome, overall ewe fertility will remain low despite improvements in the consistency of uterine AI pregnancy rates. Cervical penetration rate was estimated without the use of actual insemination to assess fertility effects. In the previous study (23), GST-AI judged to be within the uterus by the author (Windsor) produced the same pregnancy rate as laparoscopic AI. Estimation of cervical penetration was therefore considered to be accurate. It has been suggested that the lower rates of cervical penetration reported for Merinos compared to North American breeds may be a function of a shorter postpartum interval at AI (6,23). However, this study found no relationship between postpartum interval and cervical penetration. Postpartum intervals of less than 12 wk were not considered, because weaning at less than 12 wk after the start of lambing is unusual in Australia. Bucknell et al (4) report a 100% cervical penetration rate in ewes inseminated 8 to 12 wk post partum (compared with 77% at 12 wk in this study), and generally higher rates in ewes inseminated less than 4 months after lambing (3). This may reflect variation between breeds or differences in operator skill. The Australian Merino industry is, in any case, based on an annual production cycle. Insemination at post partum intervals of less than 26 wk is therefore impractical. The lack of interaction between ewe and penetration attempt indicates that penetration rate is influenced by individual ewe variation rather than random factors. Fukui and Roberts (9) also found cervical penetrability to be highly repeatable in Merino ewes (over 4 d) and suggested the possibility of a preliminary screening to select those ewes which are suitable for transcervical insemination. The majority of selection pressure is applied via the male rather than the female in most Merino breeding programs. Such ewe selection strategies would not reduce the value of GST-AI to the Australian sheep industry, unless they increased generation interval by excluding younger ewes. Previous reports have indicated that maiden ewes are not suitable for transcervical insemination (13). The present study shows a clear effect of ewe parity on the success of GSTAI in horn-tonally synchronized ewes during the nonbreeding season. This is not surprising, given the extensive cervical stretching which occurs at each lambing event. The data suggest that in some flocks and breeding systems it may be wise to exclude ewes which have lambed only once as well as maidens from transceivical AI programs. Results previously reported for GST-AI have been obtained in ewes with estrus cycles synchronized by progestagen sponges. This study showed no difference in penetration rates between ewes synchronized by progestagen or prostaglandin treatments. It also suggests the possibility that there may be advantages to inseminating at a natural, rather than a
1017
Theriogenology
synchronized, estrus. Other studies (5) suggest that the ease of cervical penetration may differ with stage of natural es&us. While hormonal synchronization may not affect penetration rates, both progestagens and prostaglandins (reviewed by Hawk; 14) have the potential to influence gamete transport in the female tract. Fertility testing will therefore be required to fully assess the relative merits of these treatments. While the Merino is less seasonal than many European breeds, a large proportion of ewes are typically anovulatory in spring and early summer (reviewed by Martin et al; 17). Comparison of the cervical penetration rates for the Experiment 2 ewes, and ewes with a similar genotype and management history from Experiment 3 suggests that the stage of the breeding season may have a large effect on cervical penetration rate. However, specifically designed experiments will be required to give a definitive answer to this question. Buckrell et al (3) reported reduced fertility in GST-AI treated ewes during the non-breeding season, but attributed this to reduced fertilization rates following uterine AI rather than to an inability to penetrate the cervix. Manipulation of the ewe and management factors described above may succeed in increasing the rates of cervical penetration in Merinos inseminated by GST-AI. In this study, different combinations of ewes and preinsemination management produced a range of cervical penetration rates from 21 to 80%. Cervical penetration rates during the breeding season compare favorably with those published elsewhere (6). Further experiments will be required to determine whether increased rates of uterine AI are matched by corresponding increases in ewe fertility. Further investigation is also required to determine why pregnancy rates achieved by uterine transcervical AI do not always match those for laparoscopic AI (6). Eppleston (5) reports that laparoscopic AI produces similar conception rates, at a range of sperm doses, whether semen is placed within the uterine body or uterine horn. This suggests that site of insemination within the uterus is unlikely to be the reason for such differences in fertility. In conclusion, this paper describes the influence of a number of factors on the cervical penetration rates achieved by GST-AI in Merino ewes. There were clear effects due to individual ewe variation and ewe parity. Penetration rates may possibly be affected by AI at a natural rather than synchronized estrus, or at different stages of the breeding season. Postpartum interval at AI or type of hormonal estrus synchronization regimen had no effect. These data suggest that ewe selection strategies and managerial changes may be combined to increase the rate of uterine insemination achieved by GST-AI programs in Merino ewes. REFERENCES 1. Andersen V, Aamdal J, Fougner J. Interuterine and deep cervical insemination with frozen semen in sheep. Zuchthygiene 1973;8:113-118. 2. Armstrong DT, Evans G. Innauterine insemination enhances fertility of tinzen semen in superovulated ewes. J Reprod Fertil 1984;71:89-94. 3. Buclcrell B, Buschbeck C, Gartley C, Kroetsch T, McCutcheon W, Martin J, Penner, Walton J. A breeding trial using a manscervical technique for artificial insemination in sheep. Proc 12th Int Cong Anim Reprod Art Insem 1992;12:1531-1533.
1018
Theriogenology
4. Buckmll BC, Buschbeck C, Gardey C, Kroetsch T, Walton J.S. Further development and testing of a tramcervical technique for AI in sheep. Theriogenology 1994;42:601-611. 5. Eppleston J. Studies on the Fertility of Frozen-Thawed Ram Semen. Ph.D. Dissertation, The University of New South Wales, Sydney, 1992. 6. Eppleston J, Maxwell WMC. Recent attempts to improve the fertility of frozen ram semen inseminated into the cervix. Wool Tech Sheep Breed 1993;41:291-302. 7. Evans G. Application of reproductive technology to the Australian livestock industries. Reprod Fertil Dev 1991;3:627-650. 8. Fukui Y, Roberts EM. Fertility of non-surgical i&a-uterine insemination with frozen-pelleted semen in ewes treated with prostaglsndin F2a In: Tomes, GJ, Robertson, DE and Lightfoot, RJ (eds) Proc Int Congr Sheep Breed, Western Australian Institute of Technology, Perth, 1976482-49. 9. Fukui Y, Roberts EM. Repeatability of non-surgical intrauterine technique for artificial insemination of the ewe. Theriogenology 1977;8:77-81. 10. Fukui Y, Roberts EM Sperm transport after non-surgical intrauterine insemination with frozen semen in ewes treated with prostaglandin F2a J Reprod Fertil1977;51:141-143. 11. Fukui Y, Roberts EM. Further studies on non-surgical interuterine technique for artificial insemination in the ewe. Theriogenology 1978;10:381-393. 12. Halbert GW, Dobson H, Walton JS, Buclaell BC. A technique for transcervical intrauterine insemination of ewes. Theriogenology 1990;33:993-1010. 13. Halbert GW, Dobson H, Walton. JS, Sharpe P, Bucktell BC. Field evaluation of a technique for transcervical intrauterine insemination of ewes. lheriogenology 1990;33:1231-1243 . 14. Hawk HW. Sperm survival and transport in the female reproductive tract. J Dairy Sci 1983;66:26452660. 15. Killeen ID, Caffery GJ. Uterine insemination of ewes with the aid of a laparoscope. Aust Vet J 1982;59:95. 16. Lightfoot RJ, Salamon S. Fertility of ram spermatozoa frozen by the pellet method. I. Transport and viability of spermatozoa within the genital tract of the ewe. J Repmd Fertll 1970;22:385-398. 17. Martin GB, Oldham CM, Cognie, Y, Peame, DT. The physiological responses of anovulatory ewes to the introduction of rams - a review. Livest Prod Sci 1986;15:219-247. 18. Maxwell WMC, Butler LG, Wilson HR. Intrauterine insemination of ewes with frozen semen. J Agric Sci 1984;102:233-235. 19. Maxwell WMC, Hewitt LJ. A comparison of vaginal, cervical and intrauterine insemination of sheep. J Agric Sci 1986;106:191-193. 20. Payne RW, Lane PW, Ainsley AE, Bicknell KE, Digby PGN, Hsrding SA, Leech PK, Simpson HR, Todd AD, Verrier PJ, White RP, Gower JP, Tunnicliffe Wilson G, Paterson LJ. Genstat 5 Reference Manual. Clamndon Press, Gxford, 1987. 21. Ritar AJ, Ball PD, G’May PJ. Artificial insemination of cashmere goats: effects on fertility and fecundity of irmavaginal treatment, method and time of insemination, semen freezing process, number of motile spermatozoa and age of females. Reprod Fertil Dev 1990,2:377-384. 22. Salamon S, Lightfoot RJ. Fertility of ram spermatozoa frozen by the pellet method. III. The effects of insemination technique, oxytocin and t&xin on lambing. J Reprod Fertil 1970;22:409-423. 23. Windsor DP, Szell AZ, Buschbeck C, Edward AY, Milton JTB, Buckrell BC. Transcervical artificial insemination of Australian Merino ewes with frozen-thawed semen. Theriogenology 1994;42:147-157.
July
20,
November
1994 10,
1994
ABSTRACT The experiments described examined the effects of a number of factors on the level of uterine insemination achieved in Merino ewes by a transcervical insemination technique (Guelph system for transcervical artificial insemination; GST-AI). Cervical penetration rate is an important limitation to the use of such methods in Merinos. Simulated insemination was performed to estimate the proportion of ewes in which a pipette could be passed through the cervix to the uterus. In Experiment 1, cervical penetration rate (n=14 to 30)was unaffected by an increase in postpartum interval at AI from 12 to 26 wk. The results of cervical penetration for individual ewes weir: found to be repeatable (PcO.05). Experiment 2 (197 ewes) revealed a clear effect of ewe parity on penetration rates in hormonally synchronized ewes during the nonbreeding season (P
Acknowledgements The Author wishes to thank T. Gardner, A. Edward and T. Ladyman for technical assistance. Thanks also to I. Dhaliwal for statistical advice and assistance, and to DI K.P. Croker for comments on the manuscript. Theriogenology 43:i 009-l 018,1995 0 1995 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010
0093-691 x/95/$1 0.00 SSDI 0093-691X(95)00065-G
1010
Theriogenology INTRODUCTION
The major factor limiting the fertility of ewes inseminated with frozen ram semen is the inability of ram spermatozoa to cross the cervix of the ewe after being subjected to the freezing and thawing processes (16). The exact nature of the cellular changes causing this problem are as yet undefined. Unlike fresh semen, which provides acceptable conception rates at cervical or vaginal insemination (19), frozen semen must be deposited within the uterus to ensure adequate fertility (22). The most widely practised form of uterine insemination in the ewe is laparoscopic AI (15). This procedure provides acceptable conception rates using both fresh and frozen semen (2,15,18,19) and is widely used within the Australian stud Merino industry (7). While this procedure is effective, its wider application is limited by its expense. An alternative to laparoscopic AI is the passage of a pipette into the uterus from die vagina via the cervix. This procedure is known as transcervical AI and, in its most familiar form, is widely practiced in cattle. Transcervical procedures are less capital intensive than laparoscopy, and do not involve surgical entry to the body cavity. The anaesthesia and veterinary skills associated with laparoscopy are not required. There have been a number of reports of successful transcervical AI in sheep (1,8-13,), but these techniques have not been widely adopted in Australia (6). One such transcervical technique was recently evaluated for use with Australian Merinos (23). That study indicated that the Guelph system for transcervical AI (GST-AI) could produce similar conception rates to laparoscopic AI, but only in those ewes in which semen was deposited within the uterus rather than the cervix (about 50% of ewes in the hands of Australian operators, 23). Consequently, cervical penetration rate (the proportion of ewes in which the AI pipette can be passed through the cervix) was identified as a major factor limiting the utility of this technique under Australian conditions. Cervical penetration rates are likely to be influenced by both operator skill and variation between ewes (reviewed by Eppleston and Maxwell; 6). One strategy to increase penetration rate may be to select only certain suitable classes of ewes for mating by transcervical insemination. Alternatively, it may be possible to manipulate pre insemination management to increase penetration rates. There is little information available on the ewe or management factors which influence cervical penetration rates by GST-AI in Merinos. This paper describes a series of experiments examining some of the factors which have the potential to influence the cervical penetration rate achieved in Merino ewes using the instruments and procedures developed for GST-AI. The data will be useful in designing experiments to test the effects of such factors on ewe fertility after GST-AI. The factors considered included postpartum interval at insemination, individual ewe variation, ewe parity, estrus synchronization technique and the stage of the breeding season.
1011
Theriogenology MATERIALS AND METHODS
Experiments 1 (60 ewes) and 2 (197 ewes) were conducted outside the normal breeding season for Merino ewes in Western Australia (August-December). Experiment 3 (161 ewes) was conducted within the breeding season (April-May). All insemination attempts were made by the same operator, who was experienced in the assessment procedure (23). All ewes were managed according to Western Australian commercial practices. Experiment 1 Mature (5 yr) Merino ewes were checked for lactational status 12 wk after the start of lambing. Ewes which had lambed were randomly allocated to two groups of 30 ewes each. The estrous cycles of the ewes were synchronized by insertion of progestagen sponges= for 12 to 14 d and the intramuscular injection of 300 IU PMSGb at the time of sponge withdrawal. The ewes were divided into two groups to avoid the need for almost continuous progestagen treatment which would have arisen had a single group been examined at three-weekly intervals. Transcervical insemination was attempted as described by Halbert et al (12) between 48 and 60 h after sponge withdrawal Ewes were restrained in dorsal recumbancy, and a tubular speculum (illuminated by a fibre optic light source) inserted into the vagina. Bozeman forceps were used to grasp and stab&e the cervix during the penetration attempt. A fine stainless steel pipette was introduced into the cervical OS,and gentle probing used to locate the passage through the cervix to the uterus. Each cervical penetration attempt was judged by the operator to have either succeeded or failed in passing the AI pipette through the cervix into the uterus (3). Cervical penetration was considered to have been successful when the AI pipette was passed around ah obstructions encountered and inserted into the cervix to its full length (8.5 cm). Because cervical penetration was the end point for this study, no semen was deposited in the reproductive tract. Cervical penetration attempts were made in Group 1 ewes at 14,20 and 26 wk after the commencement of lambing. Cervical penetration attempts were made in Group 2 ewes at 17 and 23 wk after the start of lambing. At the first time point for each group (14 and 17 wk for Groups 1 and 2, respectively), cervical penetration was attempted 3 times for each ewe to examine the repeatability of penetration (r&O). Ewes were released from the cradle and returned to their group between attempts. Ewes were classified into 8 classes (see Table 1) based on the success or failure of each penetration attempt. Fewer Group 1 ewes were available later in the experiment due to misidentification of animals following shearing.
e3Omg FGA, Ovakron, ESP, Johannesburg South Africa. bPregnecol, Horizon, North Ryde NSW.
1012
Theriogenology
Experiment 2 Ewes from a flock for which individual lambing histories were known were allocated to experimental groups on the basis of having lambed once (n=76), twice (t&8) or three times (n=53). Only ewes which had lambed at each opportunity were used to avoid any interactions between ewe parity and age. Postpartum interval for these ewes was about 8 mo. Estrus synchronization and simulated transcervical insemination were performed as in Experiment 1. Cervical penetration was assessed once for each ewe. Experiment 3 Cervical penetration in mature ewes (largely from the same flock as the ewes used in Experiment 2) with 1 or 2 lambings and a postpartum interval of about 13 mo, was assessed in ewes, with estrus controlled as follows: 1) Estrus was synchronized using progestagen sponges and PMSG as described for Experiment 1. Cervical passage was attempted 51 to 55 h after sponge withdrawal (n=51). 2) Estrus synchronized by two treatments (intramuscular injections 13 d apart) with synthetic prostaglandin F2_ (125pg cloprostenol)c and 300 IU PMSG at the second treatment. Cervical passage was attempted 51 to 55 hr after the second prostaglandin treatment (n=50). 3) Natural estrus was detected by harnessed vasectomized rams. Ewes were checked for raddle marks twice daily to detect the onset of estrus. Cervical passage was attempted 8 to 16 hr after fit estrus detection for each ewe (n=60). Statistical Analyses The proportions of ewes in which cervical penetration was successful for each treatment were compared by logistic regression using Genstat 5 (20). Separate models were fitted to examine the effects of postpartum interval, ewe parity, estrus synchronization treatment and stage of the breeding season on cervical penetration, and to assess the repeatability of cervical penetration. RESULTS Postpartum Interval The cervical penetration rate (expressed as ewes penetrated as a proportion of ewes attempted) for the first penetration attempt at each time point in Experiment 1 is shown in Figure 1. Logistic regression did not indicate any significant difference between penetration rates at the different times after the start of lambing (P=O.7). Individual Ewe Variation The number of ewes falling into each penetration class is shown in Table 1. More than 75% of ewes produced the same penetration result (1 signifying success and 0 failure) at each of 3 CEstrumate, Pitman Moore, North Ryde NSW.
1013
Theriogenology
attempts. Logistic regression showed no interaction between individual ewe and the result at each penetration attempt, indicating that the first result obtained for any ewe is repeatable (P
20
17
23
26
29
Postpartum interval (weeks)
Figure 1. Cervical penetration rates in 5 yr Merino ewes with increasing postpartum interval. Estrus was synchronized using sponges and PMSG during the non-breeding season. Postpartum interval did not affect penetration rate.
Table 1. Repeatability of cervical penetration in progestagen synchronized Merino ewes during the nonbreeding season. Penetration class represents the combination of success (1) or failure (0) at each of 3 attempts (n=60). Penetration at each of 3 attempts 000 001 010 011 100 101 110 111 Total
Number of ewes
%
12 0 4 1 3 2 4 34 60
20.0 0.0 6.7 1.7 5.0 3.3 6.7 56.6 100.0
1014
Theriogenology
Ewe Parity The cervical penetration rate in ewes having lambed once, twice or three times (Experiment 2) is shown in Figure 2. The rate of cervical passage improved with incmasing parity (P~0.05).
1
2
3
Ewe parity ( number of lambings)
Figure 2. The effect of ewe parity on cervical penetration rate in Merino ewes synchronized with sponges and PMSG in the non-breeding season. Penetration rates differed with ewe parity (P
1015
Theriogenology
Pr;g;;;gn
Pr”+“tafan~in Natural Estrus
Estrus synchronization treatment
Figure 3. Cervical penetration rates in Merino ewes synchronized using progestagen or prostaglandin, or at natural estrus during the breeding season. Penetration rates did not differ between treatments (P=O.O8).
(10/12)
Ewe age 3 years 0
4 years (28168)
(16176)
Breeding
Season
NOIItIreedlng
season
Figure 4. Cervical penetration rates in 3- and 4-year old ewes from Experiment 2 (nonbreeding season), and hormonally treated ewes originating from the same flock and used in Experiment 3 (breeding season). Penetration rate differed with season (PcO.05).
1016
Theriogenology DISCUSSION
Previous research has identified 2 major factors limiting the fertility of Merino ewes that are inseminated by GST-AI (23). These were failure to penetrate the cervix and perform a uterine insemination iu a large proportion of ewes, and the inability to consistently produce equivalent fertility to laparoscopic AI after successful transcervical AI. Similar results have been mported for other ewe breeds (3), and the goat doe (21). This present study considered only the first of these 2 problems. If poor cervical penetration rates cannot be overcome, overall ewe fertility will remain low despite improvements in the consistency of uterine AI pregnancy rates. Cervical penetration rate was estimated without the use of actual insemination to assess fertility effects. In the previous study (23), GST-AI judged to be within the uterus by the author (Windsor) produced the same pregnancy rate as laparoscopic AI. Estimation of cervical penetration was therefore considered to be accurate. It has been suggested that the lower rates of cervical penetration reported for Merinos compared to North American breeds may be a function of a shorter postpartum interval at AI (6,23). However, this study found no relationship between postpartum interval and cervical penetration. Postpartum intervals of less than 12 wk were not considered, because weaning at less than 12 wk after the start of lambing is unusual in Australia. Bucknell et al (4) report a 100% cervical penetration rate in ewes inseminated 8 to 12 wk post partum (compared with 77% at 12 wk in this study), and generally higher rates in ewes inseminated less than 4 months after lambing (3). This may reflect variation between breeds or differences in operator skill. The Australian Merino industry is, in any case, based on an annual production cycle. Insemination at post partum intervals of less than 26 wk is therefore impractical. The lack of interaction between ewe and penetration attempt indicates that penetration rate is influenced by individual ewe variation rather than random factors. Fukui and Roberts (9) also found cervical penetrability to be highly repeatable in Merino ewes (over 4 d) and suggested the possibility of a preliminary screening to select those ewes which are suitable for transcervical insemination. The majority of selection pressure is applied via the male rather than the female in most Merino breeding programs. Such ewe selection strategies would not reduce the value of GST-AI to the Australian sheep industry, unless they increased generation interval by excluding younger ewes. Previous reports have indicated that maiden ewes are not suitable for transcervical insemination (13). The present study shows a clear effect of ewe parity on the success of GSTAI in horn-tonally synchronized ewes during the nonbreeding season. This is not surprising, given the extensive cervical stretching which occurs at each lambing event. The data suggest that in some flocks and breeding systems it may be wise to exclude ewes which have lambed only once as well as maidens from transceivical AI programs. Results previously reported for GST-AI have been obtained in ewes with estrus cycles synchronized by progestagen sponges. This study showed no difference in penetration rates between ewes synchronized by progestagen or prostaglandin treatments. It also suggests the possibility that there may be advantages to inseminating at a natural, rather than a
1017
Theriogenology
synchronized, estrus. Other studies (5) suggest that the ease of cervical penetration may differ with stage of natural es&us. While hormonal synchronization may not affect penetration rates, both progestagens and prostaglandins (reviewed by Hawk; 14) have the potential to influence gamete transport in the female tract. Fertility testing will therefore be required to fully assess the relative merits of these treatments. While the Merino is less seasonal than many European breeds, a large proportion of ewes are typically anovulatory in spring and early summer (reviewed by Martin et al; 17). Comparison of the cervical penetration rates for the Experiment 2 ewes, and ewes with a similar genotype and management history from Experiment 3 suggests that the stage of the breeding season may have a large effect on cervical penetration rate. However, specifically designed experiments will be required to give a definitive answer to this question. Buckrell et al (3) reported reduced fertility in GST-AI treated ewes during the non-breeding season, but attributed this to reduced fertilization rates following uterine AI rather than to an inability to penetrate the cervix. Manipulation of the ewe and management factors described above may succeed in increasing the rates of cervical penetration in Merinos inseminated by GST-AI. In this study, different combinations of ewes and preinsemination management produced a range of cervical penetration rates from 21 to 80%. Cervical penetration rates during the breeding season compare favorably with those published elsewhere (6). Further experiments will be required to determine whether increased rates of uterine AI are matched by corresponding increases in ewe fertility. Further investigation is also required to determine why pregnancy rates achieved by uterine transcervical AI do not always match those for laparoscopic AI (6). Eppleston (5) reports that laparoscopic AI produces similar conception rates, at a range of sperm doses, whether semen is placed within the uterine body or uterine horn. This suggests that site of insemination within the uterus is unlikely to be the reason for such differences in fertility. In conclusion, this paper describes the influence of a number of factors on the cervical penetration rates achieved by GST-AI in Merino ewes. There were clear effects due to individual ewe variation and ewe parity. Penetration rates may possibly be affected by AI at a natural rather than synchronized estrus, or at different stages of the breeding season. Postpartum interval at AI or type of hormonal estrus synchronization regimen had no effect. These data suggest that ewe selection strategies and managerial changes may be combined to increase the rate of uterine insemination achieved by GST-AI programs in Merino ewes. REFERENCES 1. Andersen V, Aamdal J, Fougner J. Interuterine and deep cervical insemination with frozen semen in sheep. Zuchthygiene 1973;8:113-118. 2. Armstrong DT, Evans G. Innauterine insemination enhances fertility of tinzen semen in superovulated ewes. J Reprod Fertil 1984;71:89-94. 3. Buclcrell B, Buschbeck C, Gartley C, Kroetsch T, McCutcheon W, Martin J, Penner, Walton J. A breeding trial using a manscervical technique for artificial insemination in sheep. Proc 12th Int Cong Anim Reprod Art Insem 1992;12:1531-1533.
1018
Theriogenology
4. Buckmll BC, Buschbeck C, Gardey C, Kroetsch T, Walton J.S. Further development and testing of a tramcervical technique for AI in sheep. Theriogenology 1994;42:601-611. 5. Eppleston J. Studies on the Fertility of Frozen-Thawed Ram Semen. Ph.D. Dissertation, The University of New South Wales, Sydney, 1992. 6. Eppleston J, Maxwell WMC. Recent attempts to improve the fertility of frozen ram semen inseminated into the cervix. Wool Tech Sheep Breed 1993;41:291-302. 7. Evans G. Application of reproductive technology to the Australian livestock industries. Reprod Fertil Dev 1991;3:627-650. 8. Fukui Y, Roberts EM. Fertility of non-surgical i&a-uterine insemination with frozen-pelleted semen in ewes treated with prostaglsndin F2a In: Tomes, GJ, Robertson, DE and Lightfoot, RJ (eds) Proc Int Congr Sheep Breed, Western Australian Institute of Technology, Perth, 1976482-49. 9. Fukui Y, Roberts EM. Repeatability of non-surgical intrauterine technique for artificial insemination of the ewe. Theriogenology 1977;8:77-81. 10. Fukui Y, Roberts EM Sperm transport after non-surgical intrauterine insemination with frozen semen in ewes treated with prostaglandin F2a J Reprod Fertil1977;51:141-143. 11. Fukui Y, Roberts EM. Further studies on non-surgical interuterine technique for artificial insemination in the ewe. Theriogenology 1978;10:381-393. 12. Halbert GW, Dobson H, Walton JS, Buclaell BC. A technique for transcervical intrauterine insemination of ewes. Theriogenology 1990;33:993-1010. 13. Halbert GW, Dobson H, Walton. JS, Sharpe P, Bucktell BC. Field evaluation of a technique for transcervical intrauterine insemination of ewes. lheriogenology 1990;33:1231-1243 . 14. Hawk HW. Sperm survival and transport in the female reproductive tract. J Dairy Sci 1983;66:26452660. 15. Killeen ID, Caffery GJ. Uterine insemination of ewes with the aid of a laparoscope. Aust Vet J 1982;59:95. 16. Lightfoot RJ, Salamon S. Fertility of ram spermatozoa frozen by the pellet method. I. Transport and viability of spermatozoa within the genital tract of the ewe. J Repmd Fertll 1970;22:385-398. 17. Martin GB, Oldham CM, Cognie, Y, Peame, DT. The physiological responses of anovulatory ewes to the introduction of rams - a review. Livest Prod Sci 1986;15:219-247. 18. Maxwell WMC, Butler LG, Wilson HR. Intrauterine insemination of ewes with frozen semen. J Agric Sci 1984;102:233-235. 19. Maxwell WMC, Hewitt LJ. A comparison of vaginal, cervical and intrauterine insemination of sheep. J Agric Sci 1986;106:191-193. 20. Payne RW, Lane PW, Ainsley AE, Bicknell KE, Digby PGN, Hsrding SA, Leech PK, Simpson HR, Todd AD, Verrier PJ, White RP, Gower JP, Tunnicliffe Wilson G, Paterson LJ. Genstat 5 Reference Manual. Clamndon Press, Gxford, 1987. 21. Ritar AJ, Ball PD, G’May PJ. Artificial insemination of cashmere goats: effects on fertility and fecundity of irmavaginal treatment, method and time of insemination, semen freezing process, number of motile spermatozoa and age of females. Reprod Fertil Dev 1990,2:377-384. 22. Salamon S, Lightfoot RJ. Fertility of ram spermatozoa frozen by the pellet method. III. The effects of insemination technique, oxytocin and t&xin on lambing. J Reprod Fertil 1970;22:409-423. 23. Windsor DP, Szell AZ, Buschbeck C, Edward AY, Milton JTB, Buckrell BC. Transcervical artificial insemination of Australian Merino ewes with frozen-thawed semen. Theriogenology 1994;42:147-157.