Theriogenology. Timed artificial insemination early in the breeding season improves the reproductive performance of suckled beef cows

Theriogenology 79 (2013) 625 632 Contents lists available at SciVerse ScienceDirect Theriogenology journal homepage: www.theriojournal.com Timed artificial insemination early in the breeding season improves the reproductive performance of suckled beef cows Manoel F. Sá Filho a, *, Luciano Penteado b, Everton L. Reis a, Tomás. A.N.P.S. Reis a, Klibs N. Galvão c, Pietro S. Baruselli a a Department of Animal Reproduction, FMVZ-USP, Sao Paulo, Sao Paulo, Brazil b Firmasa, Tecnologia Para Pecuaria, Campo Grande, Mato Grosso do Sul, Brazil c Department of Animal Sciences, University of Florida, Gainesville, Florida, USA article info abstract Article history: Received 21 April 2012 Received in revised form 18 November 2012 Accepted 18 November 2012 Keywords: Bos indicus Estrus synchronization Timed artificial insemination Pregnancy Cattle The objective was to compare reproductive performance of breeding programs that used natural service (NS), AI after estrus detection (ED), and timed AI (TAI). In experiment 1, 597 suckled beef cows were randomly allocated to one of four groups. Cows in the TAIþNS group (N ¼ 150) were bred by TAI at 11 days after the onset of the breeding season (BS). Bulls were placed with cows 10 days after TAI and remained together until the end of the 90-day BS. Cows in the TAIþEDþNS group (N ¼ 148) received TAI, then AI based on ED for the next 45 days, and finally NS for the last 45 days of the BS. Cows in the EDþNS group (N ¼ 147) received AI based on ED during the first 45 days of the BS, followed by NS for the last 45 days of the BS. Cows in the NS group (N ¼ 149) were bred by NS for the entire 90-day BS. Cows in the EDþNS or NS groups had a decreased (P < 0.001) hazard of pregnancy compared with cows in the two groups bred by TAI at the onset of BS. Also, cows bred by TAI (TAIþNS ¼ 92.7%; and TAIþEDþNS ¼ 91.9%) had higher (P < 0.01) pregnancy rates at the end of the BS compared with cows not bred by TAI (EDþNS ¼ 85.0%; NS ¼ 83.2%). In experiment 2, 507 suckled beef cows were randomly assigned to one of two groups at the onset of a 90-day BS. The NS group (N ¼ 255) received only NS during the entire BS, and the TAIþNS group (N ¼ 252) received TAI at the onset of the BS, followed by NS until the end of BS. Cows in the TAIþNS group had 63% higher hazard of pregnancy (P < 0.001) compared with cows in the NS group, and reduced the median days to pregnancy by 44 (11 vs. 55 days). However, there was no difference (P ¼ 0.31) in proportion of pregnant cows at the end of the BS (TAIþNS ¼ 77.0% vs. NS ¼ 71.0%). Therefore, incorporation of TAI programs early in the BS increased reproductive performance of suckled beef cows. Ó 2013 Elsevier Inc. Open access under the Elsevier OA license. 1. Introduction The use of a defined breeding season (BS) is a common reproductive management strategy in beef cattle. This approach is used to facilitate calving and calf management, and to allow calving and breeding during optimal climatic * Corresponding author. Tel.: þ55 (11) 3091 7674; fax: þ55 (11) 3091 7674. E-mail address: manoelsa@usp.br (M.F. Sá Filho). conditions and forage availability. In tropical countries, it is common to breed during the spring and summer months (October to February) when there is higher availability of forage. Consequently, calving occurs in the spring (August to December) when it is drier, which results in lower parasite and infectious disease challenges for calves. For pasture-based systems, high pregnancy rates in the beginning of the BS are critical for herd profitability. Cows that become pregnant earlier in the BS will calve earlier in the next calving season, with additional time to recover 0093-691X Ó 2013 Elsevier Inc. Open access under the Elsevier OA license. http://dx.doi.org/10.1016/j.theriogenology.2012.11.016

626 M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 before the next BS, thereby improving the probability of re-establishing pregnancy and reducing the risk of involuntary culling [1]. Furthermore, calves born early in the calving season would be heavier at weaning, improving profitability [2,3]. In tropical climates, most of the beef cattle herds are composed of Bos indicus or crosses between Bos indicus and Bos taurus. It is noteworthy that Bos indicus cattle have a longer gestation than Bos taurus cattle (293 vs. 282 days) [4], and a longer postpartum anestrus when kept on pasture [4,5]. Delays in resumption of cyclicity are associated with an increased interval from calving to conception, reduced pregnancy rates, and economic losses [6]. To maintain a 365-day calving interval and to improve production efficiency, Bos indicus cows must conceive, on average, by 72 days postpartum. Therefore, reproductive programs for beef herds under pasture conditions should focus on increased pregnancy rates at the beginning of the BS. Artificial insemination (AI) promotes genetic and economic gains through the use of genetically superior bulls. Implementation of AI programs based on estrus detection in suckled beef herds is hampered by postpartum anestrus, estrus detection (ED) failure, large farms, many animals per lot, and labor costs [6]. Timed AI (TAI) programs allow insemination of cows regardless of cyclic status and might eliminate the need for ED. These programs also provide a systematic approach to the use of AI, facilitating its use in beef herds [6,7]. Timed AI programs allow insemination of all cows at the beginning of the BS; therefore, it is reasonable to hypothesize that it increases the proportion of pregnant cows early in the BS and consequently the overall proportion of cows pregnant the end of the BS. Although there are limited data for beef cows, there are numerous reports in dairy cows that support our hypothesis [8 10]. The objective of the present study was to evaluate the use of TAI on reproductive performance of suckled beef cows under pasture conditions. In experiment 1, reproductive performance of suckled beef cows under various reproductive strategies (using TAI, ED, and natural service [NS]) were evaluated during a 90-day BS. In experiment 2, the use of TAI at the onset of the BS, followed by NS, was compared with only NS throughout the entire 90-day BS. 2. Materials and methods 2.1. Experiment 1. Various breeding strategies in suckled beef cows 2.1.1. Cows and management All procedures were approved by the University of Sao Paulo Institutional Animal Care and Use Committee (Number 2426/2011). This experiment was conducted at a commercial beef farm located in Ribas do Rio Pardo, MS, Brazil during the 2004 to 2005 spring/summer (October to December) 90-day BS. A total of 597 suckled Nelore (Bos indicus) beef cows were enrolled. Cows were maintained on Brachiaria brizantha or Braquiaria decumbens with ad libitum access to water and mineral supplement. Cows were pastured together until onset of the BS. At onset of the BS, each experimental group was put in neighboring pastures with similar forage quality and availability. 2.1.2. General reproductive management Suckled beef cows between 55 and 75 days postpartum were randomly assigned to one of four experimental groups (TAIþNS, TAIþEDþNS, EDþNS, NS). Cows in the TAIþNS (N ¼ 150) and TAIþEDþNS (N ¼ 148) were enrolled in the TAI protocol at the onset of the BS (Day 0; Fig.1). The TAI protocol consisted of insertion of an ear implant containing 3.0 mg of norgestomet plus 3.0 mg norgestomet im, and 5.0 mg estradiol valerate (Crestar, Intervet-Schering Plough, Boxmeer, Netherlands) on Day 0. On Day 9, the implant was removed and cows were given 400 IU im equine chorionic gonadotropin (ecg; Folligon, Intervet-Schering Plough). At 52 to 56 hours after implant removal, cows were bred by TAI and were concurrently given 100 mg of GnRH (Fertagyl, Intervet-Schering Plough) [11]. Bulls were introduced to cows in the TAIþNS group 10 days after TAI. Cows in the TAIþEDþNS received TAI as described, then were observed for estrus twice daily, with AI 12 hours after ED during the first 45 days of the BS, followed by NS until the end of BS. Cows in the EDþNS (N ¼ 147) group were observed for estrus twice daily and AI was performed 12 hours after ED during the first 45 days of the BS, followed by NS until the end of the BS. Cows in the NS group (N ¼ 149) were exposed to bulls from Days 0 to 90 of the BS. 2.1.3. Management of bulls A total of 28 Nelore bulls, aged 3 to 5 years, were maintained in grazing conditions with ad libitum access to water and mineral supplementation. Bulls were tested and confirmed free of brucellosis and tuberculosis. Every bull underwent a breeding soundness evaluation 30 days before the onset of the BS, following the guidelines of the Brazilian College of Animal Reproduction [12]. Bulls were not used for NS during this prebreeding period. The breeding soundness evaluation included a physical examination, testicular evaluation, measurement of scrotal circumference, collection of a semen sample by electroejaculation, and evaluation of sperm motility and morphology. Bulls were selected based on general health aspect, absence of any abnormality of the reproductive system, and a spermiogram exceeding the minimum requirements of the Brazilian College of Animal Reproduction, namely 70% progressive sperm motility, category 3 for sperm vigor, and a maximum of 30% total abnormal sperm cells [12]. Only bulls classified as potential satisfactory breeders were used. Semen was collected from the bulls via electroejaculation and placed in a prewarmed, graded, conical plastic tube, protected (by a polystyrene cover) from light, cold shock, and rapid temperature changes. Immediately after collection, semen was maintained in a water bath at 37 C, and the proportion of progressively motile sperm (0 100%) was assessed. For this, a small drop of semen was placed on a prewarmed slide, covered with a cover slip, and examined with a bright-field microscope (magnification 400; Olympus DX50, Olympus Corporation, Tokyo, Japan) with a heated stage. The proportion of sperm that were progressively motile was estimated in increments of 5%. Concurrently, an aliquot was fixed in buffered isotonic

M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 627 Group TAI + NS (N = 150 cows) Natural service Group TAI + ED + NS (N = 148 cows) Natural service Group ED + NS (N = 147 cows) Calving Estrus detection + AI Natural service Group NS (N = 149 cows) Natural service Days of breeding season 0 9 21 45 90 Fig. 1. Schedule of experiment 1. Cows were randomly assigned to one of four experimental groups (TAIþNS, TAIþEDþNS, EDþNS, and NS). Cows in the TAIþNS and TAIþEDþNS groups received TAI early in the BS. The TAI protocol consisted of an ear implant (3.0 mg Norgestomet) plus a concurrent injection of 5.0 mg estradiol valerate and 3.0 mg norgestomet (EVþN) on the first day of the treatment. Nine days later, the ear implant was withdrawn and cows were given 400 IU ecg. Cows were bred by TAI 52 to 56 hours after ear implant removal and concurrently given 100 ug of GnRH. Cows from the TAIþNS group received TAI on Day 11 of the BS, followed by NS until the end of the BS; TAIþEDþNS: cows in the TAIþEDþNS received TAI on Day 11, then were observed for estrus twice a day with AI 12 hours after ED until Day 45 of the BS, followed by NS until the end of BS; EDþNS: cows were artificially inseminated after estrous detection during the first 45 days of the BS followed by NS until the end of the BS; NS: breeding was exclusively by NS throughout the entire BS. BS, breeding season; ED, estrus detection; NS, natural service; TAI, timed AI. formal-saline (1:200) for subsequent analysis of sperm morphology. Under phase-contrast microscopy (magnification 1250), at least 200 sperm per ejaculate were examined in random fields [13]. Sperm defects were categorized as either minor or major, and their sum as total defects [14]. Bulls were classified according to their individual breeding soundness evaluation and were equally assigned to each experimental group (7 per group) at a ratio of one bull for every 21 or 22 cows. In the NS group, bulls were introduced on the first day of the BS and remained with the cows during the entire BS. Cows from TAIþNS group were exposed to bulls 10 days after the TAI (i.e., 21 days after the start of the BS), and cows in the TAIþEDþNS and EDþNS were exposed to bulls beginning on Day 45 of the BS (Fig. 1). 2.2. Experiment 2. Effect of TAI at the beginning of the breeding season in suckled beef cows 2.2.1. Cows and management This experiment was conducted in a commercial beef farm located in Cascavel, PR, Brazil during the 2007 to 2008 spring/summer (November to January) 90-day BS. A total of 507 suckled beef cows (Nelore [Bos indicus; N ¼ 302] and crossbred [Bos taurus Bos indicus; 50% Angus 50% Nelore or 25% Angus 75% Nelore; N ¼ 205]) were enrolled. Cows were maintained on four Cynodon ssp. or Panicum maximum pastures with ad libitum access to water and mineral supplement. On the first day of the BS, body condition scores (BCS) were recorded using a 1 to 5 scale (1 ¼ emaciated, 5 ¼ obese); [15]. 2.2.2. General reproductive management Monthly, cohorts of suckled beef cows (cohort of cows that had calved at approximately the same time), between 30 and 60 days postpartum, were blocked by parity and breed (Nelore or crossbred), and within each block were randomly assigned to one of two treatments: TAIþNS (N ¼ 252) in which cows received TAI early in the BS (Day 11) followed by NS from Days 16 to 90, or NS (N ¼ 255) in which cows were bred by NS from BS Days 0 to 9, then bulls were removed for 7 days and then present from Days 16 to 90 (Fig. 2). This experimental design, with cows exposed to bulls from the first day of the BS, provided an opportunity for breeding for cows in the NS group that were in estrus. Cows in the TAIþNS group were synchronized with an estradiol plus progestin-based TAI protocol, as described in experiment 1, initiated at the onset of the BS. Throughout the BS, each breeding group was pastured together and exposed to the same group of bulls. 2.2.3. Management of bulls Nelore bulls (N ¼ 20), 3 to 5 years old, were used. The same bull management was applied, regardless of farm location or breeding group. Bulls were negative for brucellosis and tuberculosis. Every bull underwent a breeding soundness evaluation 30 days before the onset of the BS, as described in experiment 1. Only bulls classified as potential satisfactory breeders were used.

628 M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 Fig. 2. Schedule of experiment 2. Suckled beef cows were randomly assigned to one of two groups (TAIþNS or NS). Cows from the NS group received only NS throughout the entire BS, and cows from the TAIþNS received TAI on Day 11 of the BS, followed by NS from Day 16 to the end of the BS. Cows from the TAIþNS group received the same TAI protocol described in experiment 1. At ear implant removal of the TAIþNS group, bulls were removed and cows remained isolated for 7 days. On Day 16, bulls were returned, and kept with the cows until the end of the BS (90 days). BS, breeding season; NS, natural service; TAI, timed AI. Bulls were placed with cows in each monthly cohort of cows on the first day of the BS. On Day 9 (day of ear implant removal for the TAIþNS group), bulls were removed and remained isolated for 7 days, but were present in the pasture with the cows on Days 16 to 90 (Fig. 2). Therefore cows from the NS group received only NS throughout the entire BS, and cows from TAIþNS received TAI on Day 11 of the BS, followed by NS to the end of the BS. Bulls were used at a ratio of one bull for every 25 cows throughout the entire BS. 2.3. Pregnancy diagnosis and reproductive efficiency In both experiments, cows were examined for pregnancy by transrectal ultrasonography 30 days after TAI or every 30 days after exposure to bulls. Pregnancy was confirmed by visualizing a heartbeat in an embryo or fetus. Duration of pregnancy was estimated based on ultrasonographic assessment of amniotic vesicle size [16]. Pregnancy per AI (P/AI) was calculated as the number of pregnant cows 30 days after AI, divided by the total number of inseminated cows. Pregnancy at 45 days and at the end of the BS was calculated as the number of pregnant cows (AI or NS) divided by the total number of cows enrolled. 2.4. Statistical analysis In experiment 1, only the fixed effect of treatment (TAIþNS; TAIþEDþNS; EDþNS; and NS) was available; therefore, P/AI and pregnancy at 45 and 90 days of the BS were evaluated using the GLIMMIX procedure, using treatment as a fixed effect and cowas a random effect in the model. The time to pregnancy was evaluated using the LIFETEST procedure. Cow was considered the experimental unit. In experiment 2, pregnancy rates at 45 and 90 days of the BS were evaluated using the GLIMMIX procedure, including the fixed effects of treatment (TAIþNS or NS), breed (Crossbred or Nelore), parity (multiparous or primiparous), BCS at the start of the BS (low < 3 or medium 3), and interactions between treatment and other covariates. Pasture within treatment was included as a random effect and cow was considered as the experimental unit. The hazard of pregnancy was analyzed by the Cox proportional hazard model using the PHREG procedure of SAS. The full model included the same fixed effects described for the GLIMMIX model, plus pasture as a fixed effect. The time variable was the interval in days from the start of the BS (Day 0) to pregnancy. Cows that were not pregnant by the end of the 90-day BS, died, or were sold, were censored. The hazard of pregnancy (adjusted hazard ratio) estimated the relative rate of pregnancy according to the explanatory variables used. Median and mean days to pregnancy were obtained from the LIFETEST procedure of SAS. Survival plots were generated with MedCalc version 9.2 (MedCalc Software, Mariakerke, Belgium). For the GLIMMIX and PHREG models, variables were removed through backward elimination based on the Wald statistics criterion when P > 0.10. Differences with P < 0.05 were considered statistically significant, and 0.05 < P 0.10 were designated as a tendency. 3. Results 3.1. Experiment 1. Various breeding strategies in suckled beef cows The P/AI at TAI did not differ (P ¼ 0.48) between the two groups (TAIþNS ¼ 50.7; 76/150 vs. TAIþEDþNS ¼ 54.3; 81/148). Time to pregnancy was similar (P ¼ 0.70) between TAIþNS and TAIþEDþNS groups (Fig. 3), but tended to be decreased (P ¼ 0.09) for cows in the NS group compared with EDþNS group (median days to pregnancy ¼ 51 vs.

M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 629 69 days). Time to pregnancy was decreased (P < 0.001) for TAIþNS and TAIþEDþNS (median days to pregnancy ¼ 11 days) compared with EDþNS and NS groups (Fig. 3). Cows subjected to TAIþNS and TAIþEDþNS had a greater (P ¼ 0.001) proportion of cows pregnant at 45 days of the BS than cows from NS or EDþNS (Table 1). Breeding by TAI (TAIþNS ¼ 92.7% and TAIþEDþNS ¼ 91.9%) resulted in a greater (P < 0.01) proportion of cows pregnant at the end of the BS than cows not bred by TAI (EDþNS ¼ 85.0%; NS ¼ 83.2%; Table 1). 3.2. Experiment 2. Effect of TAI at the beginning of the breeding season in suckled beef cows The overall P/AI after TAI was 52.4% (Table 2). There were no significant interactions between treatment and parity (P ¼ 0.21), breed (P ¼ 0.83), or BCS (P ¼ 0.79). Cows subjected to TAIþNS had a greater (P ¼ 0.001) proportion of cows pregnant at 45 days of the BS than cows exposed only to NS; however, the proportion of cows pregnant at end of the BS did not differ (P ¼ 0.31) between treatments. Multiparous cows had a greater (P < 0.01) P/AI, and a greater (P < 0.01) proportion of pregnant cows at 45 days and at the end of the BS than primiparous cows (Table 2). Cows with BCS 3.0 had greater (P < 0.01) P/AI, and also a greater (P < 0.01) proportion of pregnant cows at 45 and 90 days of the BS than cows with a BCS < 3.0 (Table 2). Crossbred cows had a greater (P < 0.01) P/AI, and also had greater proportion of pregnant cows at 45 days and at end of the BS than Nelore cows (P 0.05; Table 2). When the entire BS was analyzed using the Cox PH model, the hazard of pregnancy was greater for TAIþNS than NS (Fig. 4); it was noteworthy that TAIþNS had fewer median days to pregnancy (Table 3). There were no Table 1 Reproductive parameters in suckled beef cows subjected to four breeding programs during a 90-day breeding season (BS). Breeding First 45 days of the BS Pregnancy during strategy a the BS Pregnancy per AI, % (N) b Service rate, % (N) c Pregnancy per AI, % (N) d 45 days, % (N) End, % (N) TAIþNS 50.7 (150) 75.3 (150) e 92.7 (150) e TAIþED þns 54.3 (148) 25.4 (67) 76.5 (17) 63.5 (148) f 91.9 (148) e EDþNS 44.0 (150) 53.0 (66) 23.3 (150) g 85.0 (147) f NS 44.3 (149) h 83.2 (149) f Abbreviations: BS, breeding season; ED, estrous detection; NS, natural service; TAI, timed AI. a Cows were subjected to four breeding strategies during a 90-day BS. TAIþNS: cows received TAI on Day 11 of the BS, followed by NS until the end of the BS; TAIþEDþNS: cows in the TAIþEDþNS group received TAI at Day 11, were observed for estrous twice daily with AI 12 hours after ED until Day 45 of the BS, followed by NS until the end of BS; EDþNS: cows received AI after ED during the first 45 days of the BS, followed by NS until the end of BS; NS: cows were bred exclusively by NS during the entire BS. b Number of pregnant cows after TAI. c Number of pregnant cows after AI after ED in the TAIþEDþNS and EDþNS groups. d Number of pregnant cows after AI after ED. e h Within a column, means without a common superscript letter differed (P < 0.05). significant interactions between treatment and parity (P ¼ 0.29), BCS (P ¼ 0.79), or breed (P ¼ 0.83). Based on Kaplan Meier survival curves, the increased hazard of pregnancy for TAIþNS was primarily because of pregnancies established by TAI at the beginning of the BS (Fig. 4). In addition to treatment, parity, BCS breed, and pasture also affected the hazard of pregnancy. Multiparous cows had a higher (P < 0.0001) hazard of pregnancy than primiparous cows (Fig. 5), which resulted in fewer median days to pregnancy (Table 3). Similarly, cows with BCS 3.0 and crossbred cows had faster (P < 0.001; Figs. 6 and 7, respectively) rate of pregnancy which resulted in fewer median days to pregnancy (Table 3). Cows in pastures 4 and 3 had a higher (P < 0.01) hazard of pregnancy than cows Table 2 Reproductive performance of suckled beef cows bred by two protocols (experiment 2). Fig. 3. Survival curves for proportion of nonpregnant cows by days of breeding season for various breeding strategies during a 90-day breeding season (experiment 1). TAIþNS (N ¼ 150): cows received TAI on Day 11 of the BS, followed by NS until the end of the BS; TAIþEDþNS (N ¼148): cows in the TAIþEDþNS received TAI at Day 11, then were observed for estrus twice daily, with AI 12 hours after ED until Day 45 of the BS, followed by NS until the end of BS; EDþNS (N ¼ 147): cows were bred by AI 12 hours after estrous detection during the first 45 days of the BS, followed by NS until the end of BS; NS (N ¼ 149): cows were bred by NS throughout the BS. BS, breeding season; ED, estrus detection; NS, natural service; TAI, timed AI. Item N P/AI, % (N) 45 days, % (N) End, % (N) Breeding strategy NS 255 46.3 (118) 71.0 (181) TAIþNS 252 52.4 (132) 63.5 (160) 77.0 (194) P 0.001 0.31 Parity Primiparous 250 41.3 (121) 36.8 (92) 58.0 (145) Multiparous 257 61.8 (131) 72.4 (186) 87.6 (225) P 0.002 <0.001 <0.001 Breed Nelore 302 45.0 (151) 46.4 (140) 62.5 (197) Crossbred 205 62.4 (101) 67.3 (138) 84.4 (173) P 0.007 0.03 0.05 BCS Low (<3) 244 40.5 (121) 38.9 (95) 58.6 (143) Medium (3) 263 62.6 (131) 69.6 (183) 86.3 (227) P <0.001 0.005 0.005 Abbreviations: BCS, body condition score; NS, natural service; P/AI, pregnancies per AI; TAI, timed AI.

630 M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 Fig. 4. Survival curves for proportion of nonpregnant cows by Day 90 of the breeding season (BS) for suckled beef cows bred by natural service (NS; dashed line; N ¼ 255) or by timed AI (TAI) at beginning of the BS followed by NS (TAIþNS; solid line; N ¼ 252) during 90-day BS (experiment 2). Median interval to pregnancy for NS and TAI groups was 55 days and 11 days (adjusted hazard ratio, 1.64; 95% confidence interval, 1.34 2.01), respectively. Fig. 5. Survival curves for proportion of nonpregnant cows by days of breeding season (BS) for multiparous (solid line; N ¼ 257) or primiparous (dashed line; N ¼ 250) suckled beef cows bred by natural service or timed AI at the beginning of the BS, followed by natural service during a 90-day BS (experiment 2). Median intervals to pregnancy for primiparous and multiparous cows were 79 and 33 days (adjusted hazard ratio, 1.99; 95% confidence interval, 1.57 2.53), respectively. in pasture 1, and cows in pasture 2 tended to have higher (P ¼ 0.06) hazard of pregnancy than cows in pasture 1(Table 3; Fig. 8). 4. Discussion The interval from onset of the BS to conception could be considered the primary parameter to evaluate the reproductive efficiency of beef farms that use a defined BS. A greater hazard of pregnancy during the first part of the BS is associated with better reproductive performance, lower risk of culling, and greater profitability [17]. The present studies clearly demonstrated that using TAI early in the breeding period increased reproductive performance of suckled beef cows maintained on pasture. In both of our experiments, cows receiving TAI had an increased rate of pregnancy and reduced days to pregnancy. Beef cattle often have prolonged postpartum anestrus (>85 days) closely related to the presence of calves and poor nutrition, especially when maintained under pasture conditions [18]. Suckled anestrous cows had an insufficient pulsatile release of LH to support the final stages of ovarian follicular development and ovulation [19,20]. Therefore, in these cows, exogenous progestins have been used to hasten resumption of cyclicity and improve fertility in TAI programs [1,7,21,22]. Furthermore, because of its FSH- and Table 3 Cox s proportional hazard model for daily pregnancy rates of suckled beef cows under pasture conditions (experiment 2). Item Number Median days to Adjusted hazard 95% CI P pregnancy ratio Breeding strategy TAIþNS 252 11 1.64 1.34 2.01 <0.0001 NS 255 55 Referent Parity Multiparous 257 33 1.99 1.57 2.53 <0.0001 Primiparous 250 79 Referent Breed Crossbred 205 33 1.43 1.15 1.77 0.001 Nelore 302 56 Referent BCS Medium 244 33 1.70 1.34 2.17 <0.0001 (3) Low (<3) 263 79 Referent Pasture 4 101 32 1.99 1.41 2.82 <0.0001 3 142 35 1.79 1.27 2.54 0.001 2 132 40 1.36 0.98 1.89 0.06 1 132 86 Referent Abbreviations: BCS, body condition score; CI, confidence interval; NS, natural service; TAI, timed AI. Fig. 6. Survival curves for proportion of nonpregnant cows by days of breeding season (BS) classified according to body condition score (BCS) as low (BCS < 3.0; dashed line; N ¼ 263) or medium (BCS 3.0; solid line; N ¼ 244) in suckled beef cows bred by natural service or timed AI at beginning of the BS, followed by natural service during a 90-day BS (experiment 2). Median intervals to pregnancy for low BCS and medium BCS cows were 79 days and 33 days (adjusted hazard ratio, 1.70; 95% confidence interval, 1.34 2.17), respectively.

M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 631 Fig. 7. Survival curves for proportion of nonpregnant cows by days of breeding season (BS) for Nelore (Bos indicus; dashed line; N ¼ 302) or crossbred (Bos taurus Bos indicus; solid line; N ¼ 205) suckled beef cows bred by natural service or timed AI at the beginning of the BS, followed by natural service during a 90-day BS (experiment 2). Median intervals to pregnancy for Nelore and Crossbred cows were 56 and 33 days (adjusted hazard ratio, 1.43; 95% confidence interval, 1.15 1.77), respectively. LH-like activities [23], ecg has improved ovarian follicular development and pregnancy rates in TAI programs for suckled beef cows, especially postpartum anestrous cows [7,11,24,25]. Clearly, the higher reproductive performance of cows bred by TAI in the present study was because of the 100% service rate (on Day 11 of the breeding season) combined with good P/AI when cows were synchronized using estradiol, a progestin, and ecg. Several studies have demonstrated positive effects of implementing TAI protocols on reproductive performance of dairy [8 10,26,27] and beef cows [28]. Treatment with intravaginal progesterone device plus prostaglandin 2a in Fig. 8. Survival curves for proportion of nonpregnant cows by days of breeding season (BS) for suckled beef cows bred by natural service or timed AI at the beginning of the BS, followed by natural service during a 90-day BS (experiment 2) pastured in pastures 1 (solid line; N ¼ 132), 2 (dashed line; N ¼ 132), 3 (dotted line; N ¼ 142), and 4 (dashed-dotted line; N ¼ 101). Median intervals to pregnancy for pastures 1, 2, 3, and 4 were 86, 40, 35, and 32 days. Adjusted hazard ratio (AHR) for 4 versus 1, 1.99; 95% confidence interval [CI], 1.41 2.82; AHR for 3 versus 1, 1.79; 95% CI, 1.27 2.53; AHR for 2 versus 1, 1.36; 95% CI, 0.98 1.89. suckled Brahman cows increased the number of cows detected in estrus, improved estrus synchronization, and first service conception rate compared with control cows [29]. Crossbred zebu cows receiving a progestin plus ecg had a shorter interval from calving to first AI, and shorter time to conception than control (untreated) cows [30,31]. Timed AI protocols have also increased reproductive performance of suckled crossbred beef cows (Bos taurus Bos indicus) by improving service and pregnancy rates, and ultimately by reducing the interval from calving to conception [28]. Also, the hazard of pregnancy during the first 45 days of the BS was greater in suckled crossbred cows receiving estradiol plus progesterone-based TAI protocols than either cows treated with Ovsynch or control untreated cows inseminated after spontaneous estrus [28]. In the present study, TAI programs at the onset of the BS ensured insemination of all cows, and improved the hazard of pregnancy at the beginning of the BS, in addition to overall reproductive efficiency of suckled beef cows under pasture conditions. In experiment 1, cows receiving TAI (TAIþNS or TAIþEDþNS) had a shorter interval to pregnancy than cows in the EDþNS or NS groups. Furthermore, cows in the NS group tended to have shorter interval to pregnancy than cows submitted to ED and NS (EDþNS). The problem of low ED has been reported by others working with suckled beef [4,28] or lactating dairy cows [8,10,32,33]. In that regard, failure in ED is a critical reason for poor reproductive efficiency of ED-based reproductive programs [34]. Although the findings of experiment 2 supported those of experiment 1, it was noteworthy that the effect of treatment was potentially confounded by an effect of pasture; therefore, results must be interpreted with some caution. In experiment 2, parity, BCS, and breed affected the reproductive performance of suckled beef cows, independent of the breeding program. Multiparous cows, cows with good BCS (3.0), and crossbred cows had greater hazard of pregnancy and a shorter interval to pregnancy than primiparous cows, cows with low BCS, or Nelore cows, respectively. Primiparous beef cows in pasture-based systems often have prolonged postpartum anestrus [20,35], and reduced P/AI to TAI programs compared with multiparous cows [36]. The energy requirements for growth, in addition to requirements for maintenance and lactation, are believed to account for the greater negative effects on cyclicity and reproductive performance of primiparous cows under pasture conditions [20,35]. Also, cows with greater BCS at the onset of the BS had a higher P/AI and greater hazard of pregnancy during the BS. The effect of BCS on fertility of beef cattle after TAI and the BS is well known [7,15,20] and physiologic associations are well documented [20,37]. Moreover, greater reproductive performance of crossbred cows might be primarily due to hybrid vigor. Nevertheless, crossbred cows were more likely than Nelore cows to have good (3.0) BCS (58.1% vs. 47.7%, P ¼ 0.02), which contributed to improved reproductive performance. 4.1. Conclusions TAI at the onset of the BS improved the proportion of suckled beef cows pregnant by AI, the hazard of pregnancy, the proportion of pregnant cows at 45 days (experiments

632 M.F. Sá Filho et al. / Theriogenology 79 (2013) 625 632 1 and 2) and the end of the 90-day BS (experiment 1). Cows receiving TAI in either experiment became pregnant sooner than cows receiving ED or NS. In addition, multiparous cows, cows having greater BCS, and crossbred cows had greater rates of pregnancy and a higher proportion of pregnant cows at the end of the BS. Incorporation of TAI programs early in the breeding season enhanced reproductive performance of suckled beef cows under pasture conditions. Acknowledgments The authors thank the staff of the Grupo Pennacchi (Cascavel-PR), Fazenda Café no Bule (Camapuã-MS) for allowing the use of their cattle and facilities during this study. This research was supported by Firmasa, Tecnologia para pecuária (Londrina, PR, Brazil) and CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior). References [1] Rhodes FM, McDougall S, Burke CR, Verkerk GA, Macmillan KL. Invited review: treatment of cows with an extended postpartum anestrous interval. J Dairy Sci 2003;86:1876 94. [2] Bó GA, Cutaia L, Chesta P, Balla E, Picinato D, Peres L, et al. Implementacion de programas de inseminación artificial en rodeos de cría de argentina, June 2005, Córdoba, Argentina, Proc VI Simposio Internacional de Reproducción Animal. Córdoba AR: IRAC; 2005. p. 97 128. [3] Cutaia L, Veneranda G, Tribulo R, Baruselli PS, Bó GA. Programas de inseminación artificial a tiempo fijo en rodeos de cría: factores que lo afectan y resultados productivos. V Simposio Internacional de Reproducción Animal. Córdoba, Argentina; 2003. p. 119 132. [4] Bó GA, Baruselli PS, Martinez MF. Pattern and manipulation of follicular development in Bos indicus cattle. Anim Reprod Sci 2003; 78:307 26. [5] Galina CS, Orihuela A, Duchateau A. Reproductive physiology in Zebu cattle. Unique reproductive aspects that affect their performance. Vet Clin North Am Food Anim Pract 1987;3:619 32. [6] Bó GA, Cutaia L, Peres LC, Pincinato D, Maraña D, Baruselli PS. Technologies for fixed-time artificial insemination and their influence on reproductive performance of Bos indicus cattle. Soc Reprod Fertil Suppl 2007;64:223 36. [7] Baruselli PS, Reis EL, Marques MO, Nasser LF, Bó GA. The use of hormonal treatments to improve reproductive performance of anestrous beef cattle in tropical climates. Anim Reprod Sci 2004; 82-83:479 86. [8] Cavestany D, Betancour H, Blanc JE, Lemaire C, Slavica J, Moreira F, et al. Reproductive efficiency in grazing lactating dairy cows under a programmed reproductive management system. Aust Vet J 2007; 85:141 7. [9] Herlihy MM, Berry DP, Crowe MA, Diskin MG, Butler ST. Evaluation of protocols to synchronize estrus and ovulation in seasonal calving pasture-based dairy production systems. J Dairy Sci 2011; 94:4488 501. [10] Gutiérrez JC, Palomares R, González R, Portillo G, Montero- Urdaneta M, Rubio-Guillén J, et al. Shortening the postpartum anoestrous interval in suckled crossbred dual purpose cows using progestagen intravaginal sponges plus ecg and PGF2a. Reprod Domest Anim 2009;44:48 54. [11] Sá Filho MF, Ayres H, Ferreira RM, Marques MO, Reis EL, Silva RCP, et al. Equine chorionic gonadotropin and gonadotropin-releasing hormone enhance fertility in a norgestomet-based, timed artificial insemination protocol in suckled Nelore (Bos indicus) cows. Theriogenology 2010;73:651 8. [12] CBRA. Manual para exame andrológico e avaliação de sêmen animal. Belo Horizonte- MG: Colégio Brasileiro de Reprodução Animal; 1998. [13] Barth AD, Oko R. Abnormal morphology of bovine spermatozoa. Ames: Iowa State University Press; 1989. [14] Blom E. The ultrastructure of some characteristics sperm defects and a proposal for a new classification of the bull spermiogram. Nord Vet Med 1973;25:383 91. [15] Ayres H, Ferreira RM, Torres-Júnior JRS, Demétrio CGB, de Lima CG, Baruselli PS. Validation of body condition score as a predictor of subcutaneous fat in Nelore (Bos indicus) cows. Livestock Science 2009;123:175 9. [16] Youngquist RS, Threlfall WR. Current therapy in large animal theriogenology. Second Edition. Philadelphia: W.B. Saunders Company; 2007. p. 298 299. [17] Morton JM. Interrelationships between herd-level reproductive performance measures based on intervals from initiation of the breeding program in year-round and seasonal calving dairy herds. J Dairy Sci 2010;93:901 10. [18] Montiel F, Ahuja C. Body condition and suckling as factors influencing the duration of postpartum anestrus in cattle: a review. Anim Reprod Sci 2005;85:1 26. [19] Williams N, Lancas M, Cameron J. Stimulation of luteinizing hormone secretion by food intake: evidence against a role for insulin. Endocrinology 1996;137:2565 71. [20] Randel RD. Nutrition and postpartum rebreeding in cattle. J Anim Sci 1990;68:853 62. [21] Odde KG. A review of synchronization of estrus in postpartum cattle. J Anim Sci 1990;68:817 30. [22] Geary TW, Whittier JC, Downing ER, LeFever DG, Silcox RW, Holland MD, et al. Pregnancy rates of postpartum beef cows that were synchronized using Syncro-Mate-B or the Ovsynch protocol. J Anim Sci 1998;76:1523 7. [23] Murphy BD, Martinuk SD. Equine chorionic gonadotropin. Endocr Rev 1991;12:27 44. [24] Sá Filho MF, Torres-Júnior JRS, Penteado L, Gimenes LU, Ferreira RM, Ayres H, et al. Equine chorionic gonadotropin improves the efficacy of a progestin-based fixed-time artificial insemination protocol in Nelore (Bos indicus) heifers. Anim Reprod Sci 2010;118:182 7. [25] Sales JNS, Crepaldi GA, Girotto RW, Souza AH, Baruselli PS. Fixed-time AI protocols replacing ecg with a single dose of FSH were less effective in stimulating follicular growth, ovulation, and fertility in suckledanestrus Nelore beef cows. Anim Reprod Sci 2011;124:12 8. [26] Lima FS, Risco CA, Thatcher MJ, Benzaquen ME, Archbald LF, Santos JEP, et al. Comparison of reproductive performance in lactating dairy cows bred by natural service or timed artificial insemination. J Dairy Sci 2009;92:5456 66. [27] Ribeiro ES, Cerri RLA, Bisinotto RS, Lima FS, Silvestre FT, Greco LF, et al. Reproductive performance of grazing dairy cows following presynchronization and resynchronization protocols. J Dairy Sci 2011;94:4984 96. [28] Baruselli PS, Marques MO, Carvalho NAT, Madureira EH, Campos Filho EP. Effect of different treatments for timed artificial insemination on the reproductive efficiency in lactating beef cows. Revista Brasileira de Reprodução Animal 2002;26:218 21. [29] Flores R, Looper ML, Kreider DL, Post NM, Rosenkrans Jr CF. Estrous behavior and initiation of estrous cycles in postpartum Brahmaninfluenced cows after treatment with progesterone and prostaglandin F2alpha. J Anim Sci 2006;84:1916 25. [30] Soto Belloso E, Portillo Martınez G, De Ondiz A, Rojas N, Soto Castillo G, Ramırez Iglesia L, et al. Improvement of reproductive performance in crossbred zebu anestrous primiparous cows by treatment with norgestomet implants or 96 h calf removal. Theriogenology 2002;57:1503 10. [31] Perea FP, De Ondiz AD, Palomares RA, Hernández HJ, González R, Soto ER. Control of postpartum anestrous with an intra-vaginal progesterone device plus ecg or calf removal for 120h in suckled crossbred cows managed in a pasture-based system. Anim Reprod Sci 2008;106:298 310. [32] Cordoba MC, Fricke PM. Evaluation of two hormonal protocols for synchronization of ovulation and timed artificial insemination in dairy cows managed in grazing-based dairies. J Dairy Sci 2001;84: 2700 8. [33] Cordoba MC, Fricke PM. Initiation of the breeding season in a grazing-based dairy by synchronization of ovulation. J Dairy Sci 2002;85:1752 63. [34] Senger PL. The estrus detection problem: new concepts, technologies, and possibilities. J Dairy Sci 1994;77:2745 53. [35] Wiltbank JN. Research needs in beef cattle reproduction. J Anim Sci 1970;31:755 62. [36] Sá Filho OG, Meneghetti M, Peres RF, Lamb GC, Vasconcelos JL. Fixed-time artificial insemination with estradiol and progesterone for Bos indicus cows II: strategies and factors affecting fertility. Theriogenology 2009;72:210 8. [37] Funston RN, Summers AF, Roberts AJ. Implications of nutritional management for beef cow-calf systems. J Anim Sci 2012;90: 2301 7.