Embryo transfer is a reproductive procedure that permits exploitation of females with high genetic value (McCue, 2011) so as to optimize animal breeding (Bülbül et al., 2010). Selk (2002) pointed out that embryo transfer generates a greater number of calves/year/female, accelerating the production of high-production beef and dairy cattle.

Embryo collections using the transcervical method involve steps that increase costs (Selk, 2002). The steps described in the bovine embryo transfer method range from donor superovulation to recipient preparation with donor (Troxel, 2013). The approximate minimum cost per embryo transfer pregnancy is estimated at $ 250 (Troxel, 2013); needless to say, it is necessary to increase productivity and reduce costs (Viana et al., 2017).

One of the stages of embryo collection is the flushing procedure, consisting of washing the uterus seven days after the first artificial insemination (Troxel, 2013). This phase involves costs related to the solution used in embryo collection (Caiado et al., 2009). In cattle, it is traditional to use Dulbecco’s phosphate-buffered saline (D-PBS) solution owing to the satisfactory responses obtained in this species (Curtis, 2015). Therefore, to improve the cost-benefit ratio, alternatives should be sought to lower costs and preserve embryos during the flushing procedure, as well as to achieve a good rate of embryos recovered per number of ovulations (Caiado et al., 2009).

Ringer’s lactate solution (RL) is exclusively used in horses for embryo collection, and there is no interference with embryo quality (Imel et al., 1981; Iuliano et al., 1985; Dippert et al., 1992; Alvarenga et al., 1993). In Brown Swiss cows, for example, Bülbül et al. (2010) used RL ($ 1.30/liter) for embryo collection and achieved 69.5% embryo recovery and mean of 6.3 collected embryos and 4.6 viable embryos per female, without evaluating traditional solution (D-PBS: $ 13.44/liter).

Against this background, the objective of this study was to determine the efficacy of RL with or without 1% fetal bovine serum (FBS) for uterine flushing of female beef cattle to recover embryos, to evaluate the quality and quantity of the recovered embryos, and to compare the recovery efficacy with that of D-PBS.

Research on animals was conducted according to the institutional committee on animal use under protocol n^o 600570317.

We collected 76 embryos from 22 animals subjected to superovulation, of which 52 were transferable, 24 had degenerated, and six were unfertilized structures. Among the transferable embryos, 75% were classified as grade 1 or 2 and 25% were grade 3, according to the Stringfellow and Seidel (1998) criteria.

The numbers of total embryos collected per group were as follows: 23 in the group with uterine flushing with RL, 21 in the group with uterine flushing with RL + 1% FBS, and 32 embryos in the group with uterine flushing with D-PBS. The numbers of transferable embryos were 13, 19, and 20 in the RL, RL + 1% FBS, and D-PBS groups, respectively.

Mean results of live weight (P = 0.552), body condition scoring (P = 0.992), number of viable embryos (P = 0.212), frozen embryos (P = 0.085), and number of total embryos (P = 0.169) obtained per female of a single superovulation showed no differences with the means used for uterine flushing (Table 3). However, there were fewer ovarian follicles in females with uterine flushing with D-PBS, evaluated before starting the superovulation protocol (P = 0.033) (Table 3).

Table 3

Variable	Solution1			P-value
	RL	RL + 1% FBS	D-PBS
Females
Number of animals	8	7	7	-
Live weight (kg)	436.88±28.27a	472.29±31.02a	507.00±69.00a	0.552
BCS (1 to 5 points)	3.75±0.13a	3.86±0.09a	3.86±0.24a	0.992
NPF (≥ 3 mm)	26.63±2.03a	19.43±2.34ab	17.71±2.12b	0.033
NCL (day of collection)	23.00±2.44a	16.43±4.33b	22.71±3.00a	0.007
Embryos
Recovery rate (%)	13.86±4.23b	15.39±4.61b	27.16±13.33a	˂0.0001
Viable embryos	1.63±0.71a	2.71±1.25a	2.86±1.39a	0.212
Unviable structures	1.25±0.49a	0.29±0.18b	1.71±1.25a	0.016
Frozen embryos	1.00±0.76a	2.00±0.95a	2.43±1.41a	0.085
Total embryos	2.88±0.85a	3.00±1.23a	4.57±1.72a	0.169

BCS - body condition scoring; NPF - number of pre-superovulation follicles; NCL - number of corpora lutea on the day of collection.

The number of corpora lutea was lower in the group of females that had uterine flushing with RL + 1% FBS (P = 0.007). The percentage of unviable structures was 43.47% in the group in which the uterine flushing was done with RL, 9.52% in those flushed with RL + 1% FBS, and 37.50% in those collected with D-PBS. The lowest percentage was observed in the RL + 1% FBS group (P = 0.016) (Table 3). The embryo recovery rate (P˂0.0001) was higher with the medium containing D-PBS, with no difference between RL and RL + 1% FBS.

The groups of animals presented similar mean weight, ranging from 436.88±28.27 to 507±69.00 kg, higher than the 400 kg found by An et al. (2016), who found similar results of body condition scoring as in the present study, which ranged from 3.75±0.13 to 3.86±0.24 points. The numbers of ovarian follicles evaluated by ultrasound before the beginning of the superovulation protocol were lower in the group with uterine flushing with D-PBS. The evaluation was performed without considering the estrous cycle; however, it was performed immediately before superovulation to analyze a possible correlation of embryo recovery. There was no significant correlation between the number of ovarian follicles analyzed in the pre-superovulation and the mean number of total embryos/female (r² = 0.2334; P = 0.296). This result may be related to the blood levels of the anti-Müllerian hormone, discussed by Rico et al. (2012) and Sevgi et al. (2019). Rico et al. (2012) collected blood samples one day before the first FSH dose administration in a protocol with progesterone and observed that the anti-Müllerian hormone did not vary during the estrous cycle. However, Sevgi et al. (2019) analyzed anti-Müllerian hormone in the 60th day after superovulation with a progesterone protocol, without considering the estrous cycle phase. The authors analyzed the follicular responses to treatments with follicle stimulating gonadotropins. Batista et al. (2014), working with Bos taurus (Holstein) heifers, found that, in those with low (13.4) and high (34.3) numbers of ovarian follicles, the anti-Müllerian hormone values ranged from 0.06 to 0.57 mg/mL in the plasma, respectively. In Bos indicus heifers (Nellore), the authors found a plasma concentration of the hormone of 0.78 mg/mL in those with 28.4 ovarian follicles and 1.20 mg/mL in those with 48.1 ovarian follicles. These authors worked with Holstein and Nellore; when the females had their ovarian follicular waves synchronized, serum samples were collected at 0, 60, and 120 days to measure anti-Müllerian hormone levels. The estrous cycle period was not considered.

In our experiment, hormone levels were not analyzed; however, it may be possible to infer that serum levels of anti-Müllerian hormone were similar in all animal groups, since, according to Rico et al. (2012) and Sevgi et al. (2019), higher concentrations of this hormone favor the production of embryos. The numbers of viable embryos in the animals of this study did not differ among the groups, with values similar to those observed by Sekizawa et al. (2012), Takahashi et al. (2013), and Facioli et al. (2020), when conducting research on the same breed of animals used in this study.

A recent study reported slightly larger total and viable embryos/female (An et al., 2016); nevertheless, these are considered lower than those observed in the European and Nellore breeds (Baruselli et al., 2006; Viana et al., 2017).

The number of unviable structures (degenerated, underdeveloped, empty pellucid zone, and/or unfertilized/female) observed in our study ranged from 0.29±0.18 to 1.71±1.25, similar to those observed by An et al. (2016) and Mapletoft et al. (2002). An et al. (2016) observed a variation ranging from 0.6 in Wagyu cattle to 1.0/female of degenerated structures using Folltropin. Mapletoft et al. (2002) showed degenerate structure variations ranging from 0.6 to 1.5/female. However, in the present study, a smaller number of unviable structures/female were observed in the group that had uterine flushing with RL + 1% FBS, the same as in those where uterine flushing was performed with RL and D-PBS (Table 3). We can discern no scientific explanation for this finding. By contrast, we used the 1% FBS to enrich the RL medium as did Alvarenga et al. (1993) in mares. Fetal bovine serum contains nutritional factors, growth factors, chelators of heavy metals (Abe and Hoshi, 2003), and vitamins (Abdel-Wahab et al., 2018). It also possesses potent antioxidant activity (Mahmoud and Nawito, 2003), many hormones, minerals, trace elements, lipids, and detoxifying factors (Gstraunthaler, 2003). Fetal bovine serum has properties that are beneficial for the embryos during the flushing procedure (Abe and Hoshi, 2003). In addition to the benefits of FBS on embryo collection, it can also be assumed that embryo production could be related to animal physiology and individuality (Monniaux et al., 2010). The major problem in commercial embryo production is the superovulatory response following treatment with gonadotropins such as FSH.

Frozen female embryos classified in grades 1 to 3 (Stringfellow and Seidel, 1998) showed no differences (P = 0.085) between the evaluated groups (Table 3), and information on the pregnancy rate is not yet available. However, performing superovulation on Wagyu females with Folltropin and freezing embryos with 1.5 M ethylene glycol solution, An et al. (2016) obtained 35.2% of pregnancy by thawing embryos and transferring them to recipients. The same authors obtained five viable frozen embryos/female from 24 donors, a response that differs from that found in the present study, in which there was a mean of 2.46 frozen embryos/female. Barros and Bényei (2000) reported that, when performing uterine flushing of Holstein-Frisian donors superovulated with FSH-P (Pluset, Serono, Rome, Italy), D-PBS, and zwitterionic medium, the means observed were of 43.15 and 36.15% of pregnancy using thawed embryos of grades 1 and 2, respectively. Barros and Bényei (2000) obtained 8.4±6.8 total structures when flushing the uterus of bovine females with zwitterionic medium and 7.8±6.3 with D-PBS.

The total number of embryos was 2.88±0.85 in females in which uterine flushing was done with RL, 3.00±1.23 in those with RL + 1% FBS, and 4.57±1.72 in those in which D-PBS was used (Table 3), resulting in a mean of 3.48 embryos/female, lower than the results obtained by An et al. (2016) of 6.2 embryos/donor, and Facioli et al. (2020) of 4.54 embryo/donor. Research from other countries studying Wagyu breed showed reduced number of embryos obtained/female, ranging from 0.20±0.03 total embryos and 0.10±0.02 viable embryos (Sekizawa et al., 2012) and total structures of 10.00±7.6 and 4.5±3.9 of viable structures (Takahashi et al., 2013).

The low embryo production (recovery rate) might be due to the Japanese black cattle breed (Wagyu) that has improved metabolism that contributes to higher marbling scores compared with other breeds. However, this may affect other aspects of the animal’s physiology, including hormone secretion and their reproductive performance, including their response to the synchronization protocols and embryo production (Facioli et al., 2020).

Results aimed at finding less expensive alternatives to D-PBS are important for Wagyu embryo transfer. Our data indicated with 95% confidence that it is possible to perform uterine flushing with RL with or without FBS, because statistically, the number of total, viable, and frozen embryos obtained in this experiment did not differ among flushing procedures. It was also the case that the group in which females had uterine flushing with RL enriched with 1% FBS had the lowest number of unviable structures among the experimental groups.

The recovery rate was higher in the females flushed with D-PBS than in those flushed with RL or RL + 1% FBS. This may have been due to the individuality of the animals as commented by Monniaux et al. (1983). According to Alvarenga et al. (1993), there was no influence of RL or D-PBS on embryo collection in mares with respect the recovery rates or pregnancies, even when these media were supplemented with 1% FBS.

It is worth noting that the media for embryo collection should have a pH varying from 7.2 to 7.6 and osmolarity around 300 mOsm (Wright Jr. and Bondioli, 1981; Contu et al., 2002). All media utilized in our experiment contained 6 g/L of NaCl in the RL and 8 g/L of NaCl in the D-PBS. According to Wright Jr. and Bondioli (1981), generally, it is suggested that the principal function of NaCl is to regulate the osmolarity of the medium, maintaining it around 300 mOsm. However, it appears that the effect of osmolarity of the media on embryonic development is minimal (Wright Jr. and Bondioli, 1981). Astiazaran and Longoria (1988) studied collection and embryo transfer in rats and cows using Hartmann’s physiological solution, which is easier to obtain than D-PBS; according to these authors, the composition of Hartmann’s medium is very similar to that of RL (Astiazaran and Longoria, 1988). According to Ayebale et al. (2017), RL is an isotonic balanced salt solution very similar to the ionic composition of the blood.

Ringer’s lactate solution or Ringer’s lactate solution + 1% fetal bovine serum are as efficient as Dulbecco’s phosphate-buffered saline solution with respect to the quality and quantity of embryos. Therefore, both solutions could be used as alternatives for embryo collection.