Pregnancy rates following timed embryo transfer with fresh or vitri ed in vitro produced embryos in lactating dairy cows under heat stress conditions

Transcription

1 Theriogenology 58 (2002) 171±182 Pregnancy rates following timed embryo transfer with fresh or vitri ed in vitro produced embryos in lactating dairy cows under heat stress conditions Y.M. Al-Katanani a, M. Drost b, R.L. Monson c, J.J. Rutledge c, C.E. Krininger III a, J. Block b, W.W. Thatcher a, P.J. Hansen a,* a Department of Animal Sciences, University of Florida, Building 499, P.O. Box , Shealy Drive, Gainesville, FL , USA b Department of Large Animal Clinical Sciences, University of Florida, Gainesville, FL , USA c Department of Animal Sciences, University of Wisconsin, Madison, WI, USA Received 6 September 2001; accepted 22 January 2002 Abstract Timed embryo transfer (TET) using in vitro produced (IVP) embryos without estrus detection can be used to reduce adverse effects of heat stress on fertility. One limitation is the poor survival of IVP embryos after cryopreservation. Objectives of this study were to con rm bene cial effects of TET on pregnancy rate during heat stress as compared to timed arti cial insemination (TAI), and to determine if cryopreservation by vitri cation could improve survival of IVP embryos transferred to dairy cattle under heat stress conditions. For vitri ed embryos (TET-V), a three-step pre-equilibration procedure was used to vitrify excellent and good quality Day 7 IVP Holstein blastocysts. For fresh IVP embryos (TET-F), Holstein oocytes were matured and fertilized; resultant embryos were cultured in modi ed KSOM for 7 days using the same method as for production of vitri ed embryos. Excellent and good quality blastocysts on Day 7 were transported to the cooperating dairy in a portable incubator. Nonpregnant, lactating Holsteins (n ˆ 155) were treated with GnRH (100 mg, i.m., Day 0), followed 7 days later by prostaglandin F 2a (PGF 2a, 25 mg, i.m.) and GnRH (100 mg) on Day 9. Cows in the TAI treatment (n ˆ 68) were inseminated the next day (Day 10) with semen from a single bull that also was used to produce embryos. Cows in the other treatments (n ˆ 33 for TET-F; n ˆ 54 for TET-V) received an embryo on Day 17 (i.e. Day 7 after anticipated ovulation and Day 8 after second GnRH treatment). The proportion of cows that responded to synchronization based on plasma progesterone concentrations on Day 10 and Day 17 was 67.7%. Pregnancy rate for all cows on Day 45 was higher * Corresponding author. Tel.: ; fax: address: hansen@animal.ufl.edu (P.J. Hansen) X/02/$ ± see front matter # 2002 Elsevier Science Inc. All rights reserved. PII: S X(02)

2 172 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 (P < 0:05) in the TET-F treatment than for the TAI and TET-V treatments (19:0 5:0, 6:2 3:6, and 6:5 4:1%). For cows responding to synchronization, pregnancy rate was also higher (P < 0:05) for TET-F than for other treatments (26:7 6:4, 5:0 4:3, and 7:4 4:7%). In the TET-F treatment group, cows producing more milk had lower (P < 0:05) pregnancy rates than cows producing less milk. In conclusion, ET of fresh IVP embryos can improve pregnancy rate under heat stress conditions, but pregnancy rate following transfer of vitri ed embryos was no better than that following TAI. # 2002 Elsevier Science Inc. All rights reserved. Keywords: Heat stress; Vitri cation; Embryo transfer; In vitro fertilization 1. Introduction There is a decrease in fertility in lactating dairy cows during summer in hot climates [1]. The magnitude of the depression depends on geographical location and milk yield [2±4]. There is little effect of air temperature on pregnancy rate of embryo transfer recipients [5] in contrast to the situation with cows subjected to arti cial insemination (AI), and pregnancy rates during heat stress are greater following embryo transfer than those obtained with AI [6±8]. Reduced seasonality of transferred embryos re ects the fact that bovine embryos become more resistant to the adverse effects of heat stress as they develop [9]. In addition, since only embryos that become morula or blastocysts are typically transferred, use of embryo transfer bypasses effects of heat stress on oocyte competence that limit embryonic development [10,11]. One limitation to widespread use of embryo transfer for increasing pregnancy rates in lactating dairy cows during heat stress is the high cost of producing embryos by superovulation or by transvaginal oocyte recovery and in vitro fertilization. In vitro produced (IVP) embryos using oocytes recovered from slaughterhouse ovaries can provide an inexpensive source of embryos. However, IVP embryos do not survive after freezing as well as do embryos produced in vivo [7,12], and no advantage over AI was seen for pregnancy rates of heat-stressed cows using cryopreserved IVP embryos [8]. The development of an ice-free cryopreservation technique called vitri cation has provided signi cant advances in embryo cryopreservation in recent years [13±16], and may prove a useful method for providing cryopreserved embryos for use under heat stress conditions. Heat stress is associated with a decreased expression of estrous behavior [17] and an increase in the percentage of undetected estrus [18]. Reduced expression of estrus could limit the number of cows which would be eligible for embryo transfer. However, one experiment [8] indicated that embryo transfer can be successfully performed without estrus detection in the summer by using the ovulation synchronization protocol called OvSynch [19,20]. There were two objectives of the current experiment. The rst was to re-evaluate timed embryo transfer (TET) as a method to improve fertility under heat stress conditions when compared to timed arti cial insemination (TAI) in lactating dairy cows. The second goal was to determine if cryopreservation by vitri cation could improve survival of IVP embryos transferred to dairy cattle under heat stress conditions.

3 2. Materials and methods 2.1. Materials All materials were purchased from Sigma (St. Louis, MO) unless otherwise noted. Modi ed Tyrode's solutions were obtained from Cell and Molecular Technologies (Lavallete, NJ) to prepare HEPES±Tyrode's albumin lactate pyruvate (TALP), IVF±TALP and sperm±talp. Heat-treated fetal calf serum (FCS) was purchased from Gibco BRL (Grand Island, NY). Percoll was obtained from Amersham Pharmacia Biotech (Uppsala, Sweden). Follicle-stimulating hormone (FSH) and luteinizing hormone (LH) were obtained from Sioux Biochemical (Sioux Center, IA). Oocyte collection medium was TCM-199 with Hank's salts without phenol red supplemented with 2% (v/v) heat-treated fetal calf serum, 0.04 USP U heparin/ml, 100 U/ml penicillin-g, 0.1 mg/ml streptomycin, and an additional 1 mm glutamine. Oocyte maturation medium was tissue culture medium-199 (BioWhittaker Inc., Walkersville, MD) with Earle's salts supplemented with 10% (v/v) heat-treated fetal calf serum, 1 mg/ml estradiol 17-b, 3 mg/ml FSH and 3 mg/ml LH, 0.2 M sodium pyruvate and 50 mg/ml gentamicin sulfate. Potassium simplex optimized medium (KSOM) was obtained from Cell and Molecular Technologies (Lavallete, NJ). The KSOM, which contains 1 mg/ml BSA, was modi ed on the day of use by adding 3 mg/ml essentially fatty acid-free BSA (EFAF-BSA), 2.5 mg/ml gentamicin, essential amino acids (basal medium Eagle) and nonessential amino acids (minimal essential medium). GnRH was Cystorelin 1 from Rhone Merieux (New York, NY). PGF 2a was Lutalyse 1 from Pharmacia-Upjohn (Kalamazoo, MI) Animals Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171± The experiment was conducted at three facilities of a commercial dairy farm located in Okeechobee County, Florida ( N, W) during August and September A total of 155 lactating dairy cows whose lactation ranged from 89 to 349 days were used. Cows were housed in three separate facilities. One facility was a free-stall barn equipped with misters and fans. Cows in the other two facilities were housed in lots in which shade structures and cooling ponds were located. Cows were fed one of several diets and were milked two times per day. Nonpregnant cows were identi ed each week and treated with PGF 2a (Lutalyse 1, 25 mg, i.m.). As per the desire of the cooperating dairy, cows showing signs of estrus were inseminated and not used for the experiment. The remaining cows were included in the experiment and subjected to an ovulation synchronization protocol 12 days after the prostaglandin treatment. Initiation of the protocol (Day 0) was timed so that cows responding to prostaglandin and not inseminated would be between Day 8 and Day 10 of the estrous cycle. Cows received intramuscular treatments of 100 mg GnRH on Day 0, 25 mg PGF 2a on Day 7, and 100 mg GnRH on Day 9. Cows were assigned at random to one of three treatments when they received the rst GnRH treatment: timed arti cial insemination (n ˆ 68), timed embryo transfer-fresh (TET-F, n ˆ 33), or timed embryo transfervitri ed (TET-V, n ˆ 54). Cows in the TAI treatment were inseminated on Day 10 (24 h

4 174 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 after the second GnRH treatment) using frozen semen. A single Holstein bull that was recommended by the participating dairy was used to inseminate all cows and to produce all embryos in vitro. Cows in the TET treatments were not inseminated but received either a fresh or vitri ed embryo on Day 17 (i.e. Day 7 after anticipated ovulation and Day 8 after the second GnRH treatment). Transfer of fresh embryos was performed every other week due to the availability of embryos, while transfer of vitri ed embryos was performed every week. Accordingly, fewer cows were included in the TET-F group as compared to TAI and TET-V groups. Blood samples were collected from all cows on the day of anticipated ovulation (Day 10) and on the day of embryo transfer (Day 17; Day 7 after anticipated ovulation). Samples were collected by coccygeal venipuncture into evacuated heparinized tubes (Becton Dickinson, Franklin Lakes, NJ), placed in an ice chest and centrifuged (2000 g for 20 min at 4 8C) upon return to the laboratory (10 h after collection). Plasma was separated and stored at 20 8C until assayed for progesterone by single-antibody radioimmunoassay [21]. The concentrations of progesterone on Day 10 and Day 17 were used to identify the proportion of cows that responded to synchronization Production of vitri ed embryos Embryos for vitri cation were produced at the University of Wisconsin using ovaries obtained from Holstein cows at a commercial slaughterhouse (Peck Packing Co., Milwaukee, WI). Procedures for oocyte collection and maturation, fertilization, and embryo culture were as described previously [15] except that modi ed KSOM was used as the medium for embryo culture. Fertilization was conducted with frozen-thawed and Percollseparated sperm [22] from the same Holstein bull that was used to inseminate cows in the TAI treatment. Heat-treated fetal calf serum (10%, v/v) was added to each drop on Day 5 after insemination (fertilization on Day 0). On Day 7 after fertilization, excellent and good quality blastocysts (primarily unexpanded) [23] were vitri ed using a three-step pre-equilibration procedure described previously [15]. The vitri cation solutions consisted of glycerol and ethylene glycol and were prepared in Dulbecco's phosphate-buffered saline (DPBS) supplemented with 15% (v/v) heat-treated fetal calf serum, 0.1 g/ml glucose, and 3.6 mg/ml sodium pyruvate. All the vitri cation procedures were performed at room temperature. In the rst step, embryos of excellent or good quality were transferred in groups of 10 to a solution of 10% (v/v) glycerol for 5 min; in the second step, embryos were transferred to a solution consisting of 10% (v/v) glycerol and 20% (v/v) ethylene glycol for 5 min; and in the last step (Step 3), embryos were transferred to a 70-ml drop of 25% glycerol and 25% ethylene glycol for 30 s. Embryos were then loaded in groups of 10 per straw into 0.25-ml French straws using a 1-ml syringe attached to a pipette tip. Embryos were located in a column of vitri cation solution that was used in Step 3 separated from two bracketing columns of 1.0 M sucrose separated by air space. The straws were sealed with a straw plug and plunged partially into liquid nitrogen in a vertical position for a period of 10 s, and then the straws were immersed completely into liquid nitrogen.

5 2.4. Production of fresh embryos Due to the availability of oocytes from the slaughterhouse, fresh embryos were produced every other week. Cumulus±oocyte complexes (COCs) were collected as described previously [15] and shipped overnight in maturation medium to Florida in a portable incubator (Minitube of America, Verona, WI) set at C. Upon arrival (24 h after collection), the COCs were washed once with HEPES±TALP and fertilized with spermatozoa prepared from semen of the same bull as used for the vitri ed embryos. Procedures for fertilization and embryo culture followed those described elsewhere [10] except that COCs were cocultured with sperm for 20 h at C and 5% (v/v) CO 2 in humidi ed air. Putative zygotes were denuded of cumulus cells and cultured in groups of 30 in pre-equilibrated modi ed KSOM medium. Fetal calf serum (10%, v/v) was added to each drop on Day 5 after fertilization. Blastocysts were harvested for transfer on Day 7 after fertilization. For this purpose, excellent and good quality blastocysts (early, expanding and expanded) were identi ed using a 40 stereo-zoom microscope and packaged in groups of 10±15 in a 5-ml sterile tube containing TL±HEPES supplemented with 3 mg/ml BSA, ph ˆ 7:2±7.6 (BioWhittaker Inc., Walkersville, MD). The tubes were sealed with para lm, loaded in a portable incubator set at C, and transported to the farm Preparation of embryos for transfer After transport to the farm, fresh embryos were evaluated again and those of excellent and good quality were loaded individually in 0.25-ml French straws. Vitri ed embryos were thawed at the farm by immersing each straw in a water bath at C for 10 s. The contents of the straw were expelled into an empty Petri dish. Embryos were transferred immediately to a 1.0 M sucrose solution in HEPES±TALP and held there for 5 min. After 5 min, embryos were transferred serially to 0.5 and 0.25 M sucrose solutions in HEPES± TALP for 5 min each. Embryos of excellent and good quality were loaded individually in a 0.25-ml straw and transferred to recipients. After the end of the trial, the same thawing procedure was used to evaluate the reexpansion and hatching rates of some vitri ed embryos (n ˆ 34 embryos) in vitro. Embryos were thawed as described earlier and cultured in pre-equilibrated 50-ml microdrops of modi ed KSOM to evaluate the re-expansion and hatching rates of some vitri ed embryos 24, 48 and 72 h after thawing Embryo transfer Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171± On Day 17, cows in all three treatments were palpated per rectum for the presence of a corpus luteum (CL) and for any abnormalities in the reproductive tract. Cows with severe adhesions were eliminated from the experiment. A blood sample was collected from all cows, and body condition score (BCS) was recorded on a scale of 1±5 with 0.25 increments [24]. Cows in the timed embryo transfer treatments (TET-F and TET-V) received an embryo after epidural anesthesia regardless of the presence or absence of a palpable CL. All cows in the TET treatments received an embryo because all cows in the TAI treatment

6 176 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 were included in the experiment unless an abnormality was observed upon palpation. An embryo was transferred nonsurgically to the uterine horn ipsilateral to the CL. Cows with no palpable CL received an embryo in the uterine horn ipsilateral to the larger ovary. The time between evaluating fresh embryos or thawing vitri ed embryos and transfer to recipients did not exceed 2 h. Pregnancy was determined by palpation per rectum on Day 45 after the second GnRH treatment by the reproductive management personnel at the dairy Statistical analysis In addition to treatment, cows were classi ed according to several categories. Based on plasma progesterone concentration on the day of anticipated ovulation (Day 10) and on the day of embryo transfer (Day 17), cows were distributed between four classes. Cows were considered to have low plasma progesterone on the day of anticipated ovulation if concentrations were 1.5 ng/ml. A progesterone concentration 1.5 ng/ml on the day of embryo transfer was considered high. The four classes based on this criterion starting with progesterone on the day of anticipated ovulation were low±low (n ˆ 23 cows; 1.5 ng/ml on the day of anticipated ovulation and 1.5 ng/ml on the day of embryo transfer), low±high (n ˆ 105 cows; those that responded to the synchronization protocol), high±low (n ˆ 9 cows) and high±high (n ˆ 18 cows). A second analysis was conducted using similar categories except that the cutoff for progesterone concentrations was 1.0 ng/ ml instead of 1.5 ng/ml. Unless otherwise stated, data on synchronized cows refer to the data set using the 1.5 ng/ml cutoff. Cows were grouped by BCS into two classes (2.5 and >2.5). Mature equivalent (ME) milk yield data as calculated by PC-DART (Dairy Records Processing Center, Raleigh, NC) were obtained for 150 cows; these data were used to group cows in three classes (Class 1: 7711 kg, n ˆ 32; Class 2: >7711 and 9979 kg, n ˆ 91; Class 3: >9979 kg, n ˆ 27). Postpartum interval (PPI) was the interval between calving and the day of anticipated ovulation (i.e. Day 10). Cows were assigned into one of two classes (Class 1: 150 days, n ˆ 78 cows; Class 2: >150 days postpartum, n ˆ 77 cows). Pregnancy data were analyzed by the least squares analysis of variance using the General Linear Models procedure of SAS [25]. A subset of data of only cows that responded to the synchronization (low progesterone on Day 10 and high progesterone on Day 17) was also analyzed. For both data sets, the initial model included the main effect of facility, week of transfer, treatment, body condition score class, milk yield class and PPI class, and all possible interactions. Data were subsequently reanalyzed after removing nonsigni cant effects. Treatment effects were partitioned into individual degree of freedom contrasts to test (a) TET-V and TAI versus TET-F and (b) TAI versus TET-V. An analysis was done to evaluate treatment effects on the proportion of cows successfully synchronized. The mathematical model included the effect of treatment, week, ME class, BCS class and PPI class as well as all possible interactions. Data were subsequently reanalyzed after removing the nonsigni cant effects. Pregnancy data for all cows and for the synchronized cows were analyzed by logistic regression analysis using the Proc Logistic procedure of SAS. The mathematical model included the effects of week, treatment, ME class, PPI class, and BCS class.

7 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171± Levels of statistical signi cance obtained by least squares analysis of variance and logistic regression were similar. Accordingly, only results obtained using analysis of variance are presented. 3. Results 3.1. Comparison of TAI with TET-F and TET-V Pregnancy rate on Day 45 after anticipated ovulation was higher (P < 0:05) in the TET- F treatment compared to TAI and TET-V treatments (Table 1). There was no difference between TET-V and TAI groups. The proportion of cows that responded to synchronization based on plasma progesterone concentration (1.5 ng/ml on the day of anticipated ovulation and 1.5 ng/ml on the day of embryo transfer) was 67.7% (105/155 cows). Among these synchronized cows, pregnancy rate was higher (P < 0:05) in cows receiving a fresh embryo (TET-F) in comparison to TAI and TET-V groups, and there was no statistical difference between TAI and TET-V groups. Results for synchronized cows changed only slightly if a value of 1 ng/ml was used to determine whether a cow was successfully synchronized. With this criterion, 65.8% of cows were successfully synchronized (102/155 cows) and least squares means S:E:M: for pregnancy rate were 5:3 4:2, 22:0 6:2, and 7:5 4:6% for TAI, TET-F, and TET-V groups, respectively (TET-F versus TAI and TET-V, P < 0:05). Using a value of 1.5 ng/ml progesterone as a selection criterion, there were four fewer pregnancies in the subset of synchronized cows as compared to the entire data set. One pregnant cow did not have a blood sample on Day 17 and was not included in the synchronized group. In addition, three cows (one TAI, one TET-F, and one TET-V) that were not synchronized, as determined by progesterone concentrations, were pregnant on Day 45. Two of these cows had high progesterone concentrations on Day 10 and Day 17 (3.1 and 5.8 ng/ml for the TAI cow, and 3.6 and 7.9 ng/ml for the TET-F cow). The third animal, a TET-V cow, had progesterone concentrations of 0.37 and 0.21 ng/ml on Day 10 and Day 17, respectively. No corpus luteum was palpated on Table 1 Pregnancy rate on Day 45 following timed artificial insemination (TAI) or timed embryo transfer using fresh (TET-F) or vitrified (TET-V) embryos Group Pregnancy rate 1 All cows (n ˆ 155) Synchronized cows 2 (n ˆ 105) TAI (5/68) a (3/46) a TET-F (6/33) b (5/20) b TET-V (3/54) a (2/39) a Different alphabets in superscript represent least squares means that differ (P < 0:05) as determined by orthogonal contrasts (TET-F vs. TET-V and TAI, TAI vs. TET-V). 1 Data are least squares means S:E:M:; in parentheses: number of pregnant cows/total number of cows. 2 Cows considered as synchronized were those with progesterone concentrations 1.5 ng/ml on the day of anticipated ovulation and 1.5 ng/ml on the day of embryo transfer.

8 178 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 Day 17. The cow was subsequently recorded as having aborted on an unspeci ed date. In all of these three cases, blood samples were assayed in two separate assays to con rm progesterone values Effect of BCS, ME, and PPI classes on pregnancy rate The least squares means S:E:M: for the BCS were 2:7 0:06, 2:6 0:07, and 2:7 0:06 for the TAI, TET-F, and TET-V treatments, respectively. In the initial analysis, pregnancy rate was higher (P < 0:05) in cows with BCS 2:5 when compared to cows with BCS > 2:5 (Table 2). This effect was seen in all three treatments: the least squares means S:E:M: for cows with BCS 2:5 were 8:2 6:4, 20:3 10:2, and 10:7 7:5 for the TAI, TET-F, and TET-V treatments, respectively. For cows with BCS > 2:5, the least squares means S:E:M: were 5:5 7:2, 6:2 8:0, and 3:9 6:5 in the TAI, TET-F, and TET-V treatments, respectively. For synchronized cows, there was also an effect (P < 0:05) of BCS on pregnancy rate. There was no interaction between BCS and treatment. There was no effect of BCS class on the proportion of cows successfully synchronized (i.e. low±high class) or of postpartum interval class on pregnancy rate or on the proportion of cows successfully synchronized. There was a trend for pregnancy rate to be lower for cows producing more milk (Table 2), but it was not statistically signi cant. The mathematical model included the effect of treatment, PPI class, and treatment PPI class interaction. Further analysis was performed where data for each group were analyzed separately. In this case, pregnancy rate in the TET-F group was higher (P < 0:05) for cows producing the least amount of milk (Table 3). The proportion of cows that responded to synchronization was also affected (P < 0:05) by milk yield class. The least squares means S:E:M: for the proportion of cows successfully synchronized (using 1.5 ng/ml progesterone as the criterion) were 76:2 8:5, 59:9 5:3, and 85:9 10:0% for milk yield Class 1, Class 2, and Class 3, respectively. Table 2 Effect of body condition score class (BCS), mature equivalent (ME) milk yield class and postpartum interval class (PPI) on pregnancy rates in cows following TAI, TET-F or TET-V Variable Variable class Pregnancy rate a P-value All cows Synchronized cows b BCS > ME (1) 7711 kg NS (2) >7711 and 9979 kg (3) >9979 kg PPI < NS NS, not signi cant. a Data represent least squares means S:E:M: b Cows considered as synchronized were those with progesterone concentrations 1.5 ng/ml on the day of anticipated ovulation and 1.5 ng/ml on the day of embryo transfer.

9 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171± Table 3 Effect of milk yield on pregnancy rate after TAI or TET-F, or TET-V Group Number of cows Milk yield class 1 Pregnancy rate 2 TAI 65 (1) 7711 kg (13) (2) >7711 and 9979 kg (42) (3) >9979 kg (10) TET-F 33 (1) 7711 kg a (10) (2) >7711 and 9979 kg b (18) (3) >9979 kg b (5) TET-V 52 (1) 7711 kg (9) (2) >7711 and 9979 kg (31) (3) >9979 kg (12) Different alphabets in superscript represent least squares means that differ (P < 0:05) as determined by orthogonal contrasts (Class 1 vs. Class 2 and Class 3, Class 2 vs. Class 3). 1 Cows were classi ed based on predicted 305-day milk yield (kg). 2 Data represent least squares means S:E:M:; in parentheses: the number of cows in each milk yield class Re-expansion of embryos The re-expansion rate of vitri ed embryos was 32.6% (14/34) at 24 h after thawing. Hatching rate was 9.3% at 72 h post-thawing. 4. Discussion Results obtained from this experiment con rm a previous observation [8] that timed embryo transfer with fresh in vitro produced embryos can improve fertility in lactating dairy cows under heat stress conditions as compared to pregnancy rates achieved with TAI. The effectiveness of this protocol is based on two factors. Firstly, TET bypasses losses due to effects of heat stress on the oocyte and early embryo, since some of the heat stressassociated decline in fertility is due to damage to the oocyte [10,11,26], or to the embryo before Day 3 after breeding [9]. A good or excellent quality embryo transferred on Day 7 has already escaped possible deleterious effects of heat stress during oocyte or early embryo development. Secondly, timed embryo transfer bypasses the high estrus nondetection rate during heat stress [18] because the high degree of synchrony achieved with the OvSynch protocol allows embryo transfer without estrus detection. The proportion of cows that were successfully synchronized based on plasma progesterone concentration in this study was 65.8±67.7%, a value consistent with the 76.2% rate reported in an earlier study [8]. The pregnancy rates following TET-F among cows successfully synchronized were higher than for all cows. Accordingly, further improvements in the effectiveness of the OvSynch protocol to increase the proportion of cows synchronized should result in a further increase in pregnancy rate. Even among cows that were successfully synchronized, the pregnancy rate was only 26.7% for the TET-F treatment, while pregnancy rate for single fresh IVP embryo transfer in other studies conducted in cooler climates varied between 42 and 53% [15,27].

10 180 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 Perhaps effects of heat stress on embryonic survival in the later stages of development [28] contributed to lower pregnancy rates reported in our study. Less embryo/recipient synchrony using TET as compared to transfer at a xed time after estrus and other differences between herds could also contribute to lower embryonic survival. Three cows were diagnosed pregnant even though analysis of progesterone concentrations suggested that cows were not successfully synchronized. Two of these cows had high progesterone concentrations on the anticipated day of ovulation and 7 days later. One cow received a fresh embryo, and it is not surprising that this cow became pregnant when the embryo was transferred at a stage when it could act to maintain corpus luteum lifespan. The other cow, however, was subjected to TAI. It is likely that ovulation occurred in this cow despite high concentrations of progesterone on the anticipated day of ovulation. The third anomaly was for a TET-V cow that had low progesterone concentrations on the day of embryo transfer that was con rmed by reanalysis of the blood sample and by the absence of a detected corpus luteum at the time the embryo was transferred. It is dif cult to see how this cow, which was recorded as having subsequently aborted after pregnancy diagnosis, became pregnant. Perhaps the original pregnancy diagnosis was wrong. Alternatively, a spontaneous ovulation soon after transfer may have provided the luteal support to establish pregnancy in this cow. Although the low number of cows used warrants caution, one cow factor that appeared to affect embryonic survival after transfer of a fresh embryo was milk yield. In particular, pregnancy rate was higher in cows with predicted 305-day milk yield less than 7711 kg than for cows producing more milk. There was no relationship between milk yield and pregnancy rates in the TAI and TET-V groups, perhaps because pregnancy rates were very low in these groups. The antagonistic relationship between milk yield and fertility has been reported in several studies based on analysis of large data sets [2,4,29]. The fact that high milk yield appeared to reduce fertility in embryo transfer recipients suggests that at least some of the effects of milk yield on fertility can occur after embryos reach the blastocyst stage. The cause of embryonic loss due to high milk yield in the present study could be due in part to endocrine differences associated with genetic ability for milk yield [30]. Also, liver blood ow and progesterone metabolism increase as feed intake increases [31], which might suggest that changes in progesterone metabolism associated with increased milk production may play a role in reduced fertility in cows with high milk production. Moreover, reduced ability of cows with high milk production to regulate body temperature during heat stress [32] may contribute to lower pregnancy rates. The relationship between milk yield and pregnancy rate after embryo transfer requires further study. While TET-F increased pregnancy rate compared to the TET-V and TAI groups, there was no improvement in pregnancy rates after transfer of vitri ed embryos. Similarly, improvements in summer pregnancy rates with embryo transfer over AI were not seen when conventionally frozen-thawed IVP embryos were used [7,8]. The fact that IVP embryos are more sensitive to cryopreservation is well known [12,33]. The decision in the current study to evaluate vitri cation was based on the nding that pregnancy rates of 35± 60% were reported in Wisconsin and Argentina following transfer of vitri ed embryos [15,16,34]. One explanation for the low pregnancy rates obtained in our study could be the fact that vitri ed embryos in the current study were also less able to survive in vitro after thawing compared to embryos in other experiments. In the Wisconsin study [15], the

11 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171± re-expansion and hatching rates were 90:0 0:26 and 76:0 0:27, respectively, whereas similar values in the present study were 32.6% re-expansion and 9.3% hatching rate. Another possibility is that heat stress experienced by lactating dairy cows in south Florida might have compromised the ability of vitri ed embryos to survive after transfer into recipients. In conclusion, these data con rm the bene cial effects of timed embryo transfer for improvement of fertility in lactating dairy cows exposed to heat stress. Further improvements in the degree of synchrony that can be achieved with ovulation synchronization protocols and development of reliable procedures for cryopreservation of IVP embryos will improve the practical use of timed embryo transfer as a reproductive management tool for lactating dairy cows under commercial conditions. Acknowledgements The authors thank Marie-Joelle Thatcher, Oscar Hernandez, Andrew Majewski, RocõÂo Rivera, Fabiola Paula-Lopes and Paolete Soto for technical assistance and John Gilliland and the staff at McArthur Dairy in Okeechobee, Florida for providing the animals and help with the experiment. This is journal series no. R of the Florida Agricultural Experiment Station. Research was supported in part by a grant from Florida Milk Check-Off Program. References [1] Hansen PJ. Effects of environment on bovine reproduction. In: Youngquist RS, editor. Current therapy in large animal theriogenology. Philadelphia: WB Saunders, p. 403±15. [2] Badinga L, Collier RJ, Wilcox CJ, Thatcher WW. Interrelationships of milk yield, body weight and reproductive performance. J Dairy Sci 1985;68:1828±31. [3] Badinga L, Collier RJ, Thatcher WW, Wilcox CJ. Effects of climatic and management factors on conception rate of dairy cattle in subtropical environment. J Dairy Sci 1985;68:78±85. [4] Al-Katanani YM, Webb DW, Hansen PJ. Factors affecting seasonal variation in 90-day non-return rate to rst service in lactating Holstein cows in a hot climate. J Dairy Sci 1999;82:2611±6. [5] Putney DJ, Thatcher WW, Drost M, Wright JM, DeLorenzo MA. In uence of environmental temperature on reproductive performance of bovine embryo donors and recipients in the southwest region of the United States. Theriogenology 1988;30:905±22. [6] Putney DJ, Drost M, Thatcher WW. In uence of summer heat stress on pregnancy rates of lactating dairy cattle following embryo transfer or arti cial insemination. Theriogenology 1989;31:765±78. [7] Drost M, Ambrose JD, Thatcher M-J, Cantrell CK, Wolfsdorf KE, Hasler JF, et al. Conception rates after arti cial insemination or embryo transfer in lactating dairy cows during summer in Florida. Theriogenology 1999;52:1161±7. [8] Ambrose JD, Drost M, Monson RL, Rutledge JJ, Leibfried-Rutledge ML, Thatcher MJ. Ef cacy of timed embryo transfer with fresh and frozen in vitro produced embryos to increase pregnancy rates in heatstressed dairy cattle. J Dairy Sci 1999;82:2369±76. [9] Ealy AD, Drost M, Hansen PJ. Developmental changes in embryonic resistance to adverse effects of maternal heat stress in cows. J Dairy Sci 1993;76:2899±905. [10] Al-Katanani YM, Paula-Lopes FF, Hansen PJ. Effect of season and exposure to heat stress on oocyte quality of Holstein cows. J Dairy Sci 2002;85:390±6.

12 182 Y.M. Al-Katanani et al. / Theriogenology 58 (2002) 171±182 [11] Zeron Y, Ocheretny A, Kedar O, Borochov A, Skla D, Arav A. Seasonal changes in bovine fertility: relation to developmental competence of oocytes, membrane properties and fatty acid composition of follicles. Reproduction 2001;121:447±54. [12] Massip A, Mermillod P, Dinnyes A. Morphology and biochemistry of in-vitro produced bovine embryos: implications for their cryopreservation. Hum Reprod 1995;10:3004±11. [13] Liu Y, Wang S, Holyoak GR, Bunch TD. Survival rates of in vitro produced bovine embryos cryopreserved by controlled slow-freezing, fast-freezing and vitri cation. Theriogenology 1996;45:177 [abstract]. [14] Dinnyes A, Carolan C, Lonergan P, Massip A, Mermillod P. Survival of frozen or vitri ed bovine blastocysts produced in vitro in synthetic oviduct uid. Theriogenology 1996;46:1425±39. [15] Agca Y, Monson RL, Northey DL, Abas Mazni O, Schaefer DM, Rutledge JJ. Transfer of fresh and cryopreserved IVP bovine embryos: normal calving birth weight and gestation length. Theriogenology 1998;50:147±62. [16] Agca Y, Monson RL, Northey DL, Peschel DE, Schaefer DM, Rutledge JJ. Normal calves from transfer of biopsied, sexed and vitri ed IVP bovine embryos. Theriogenology 1998;50:129±45. [17] Nebel RL, Jobst SM, Drans eld MBG, Pandol SM, Bailey TL. Use of radio frequency data communication system HeatWatch to describe behavioral estrus in dairy cattle. J Dairy Sci 1997;80(Suppl 1):179 [abstract]. [18] Thatcher WW, Collier RJ. Effects of climate on bovine reproduction. In: Morrow DA, editor. Current therapy in theriogenology. Philadelphia: WB Saunders, p. 301±9. [19] Pursley JR, Mee MO, Wiltbank MC. Synchronization of ovulation in dairy cows using PGF 2a and GnRH. Theriogenology 1995;44:915±23. [20] Schmitt EJP, Diaz T, Drost M, Thatcher WW. Use of a gonadotropin-releasing hormone agonist or human chorionic gonadotropin for timed insemination in cattle. J Anim Sci 1996;74:1084±91. [21] Knickerbocker JJ, Thatcher WW, Bazer FW, Drost M, Barron DH, Fincher KB, et al. Proteins secreted by day 16 to 18 conceptuses extended corpus luteum function in cows. J Reprod Fertil 1986;77:381±91. [22] Parrish JJ, Susko-Parrish JL, Leibfried-Rutledge ML, Critser ES, Eyestone WH, First NL. Bovine in vitro fertilization with frozen-thawed semen. Theriogenology 1986;25:591±600. [23] Stringfellow DA, Seidel SM. Manual of the International Embryo Transfer Society, Savoy, IL, [24] Edmonson AJ, Lean IJ, Weaver DL, Farver T, Webster G. A body condition scoring chart for Holstein dairy cows. J Dairy Sci 1989;72:68±78. [25] SAS. Statistical analysis system: a user's guide, version 6. Cary, NC: Statistical Analysis Systems Institute, Inc., [26] Putney DJ, Mullins S, Thatcher WW, Drost M, Gross TS. Embryonic development in superovulated dairy cattle exposed to elevated temperatures between the onset of estrus and insemination. Anim Reprod Sci 1989;19:37±51. [27] Hasler JF, Henderson WB, Hurtgen PJ, Jin ZQ, McCauley AD, Mower SA, et al. Production, freezing and transfer of bovine IVF embryos and subsequent calving results. Theriogenology 1995;43:141±52. [28] Biggers BG, Geisert RD, Wetteman RP, Buchanan DS. Effect of heat stress on early embryonic development in the beef cow. J Anim Sci 1987;64:1512±8. [29] Dematawewa CM, Berger PJ. Genetic and phenotypic parameters for 305-day yield, fertility, and survival in Holsteins. J Dairy Sci 1998;81:2700±9. [30] Lucy MC, Weber WJ, Baumgard LH, Seguin BS, Koenigsfeld AT, Hansen LB, et al. Reproductive endocrinology of lactating dairy cows selected for increased milk production. J Dairy Sci 1998;81(Suppl 1):246 [abstract]. [31] Sangsritavong S, Combs DK, Sartori RF, Wiltbank MC. Liver blood ow and steroid metabolism are increased by both acute feeding and hypertrophy of the digestive tract. J Anim Sci 2000;78(Suppl 1):221 [abstract]. [32] Berman A, Folman Y, Kaim M, Mamen M, Herz Z, Wolfenson D. Upper critical temperatures and forced ventilation effects for high-yielding dairy cows in a subtropical climate. J Dairy Sci 1985;68:1488±95. [33] Massip A. Cryopreservation of embryos of farm animals. Reprod Dom Anim 2001;36:49±55. [34] Martinez AG, de Matos DG, Furnus CC, Brogliatti GM. In vitro evaluation and pregnancy rates after vitri cation of in vitro produced bovine embryos. Theriogenology 1998;50:757±67.