Protected and partially protected fat sources from ruminal degradation for pregnant sheep

Fat sources are important component on animal feed. However, ruminants did not evolve to efficient degrade lipids. Thus, it is necessary to study how its supplementation may affect animal performance. The objective of this study was to evaluate the use of different lipids sources over consumption, performance, metabolites, and feed behavior of pregnant sheep. The experiment was carried out at Capim Branco Experimental Farm, at goat and sheep sector of Federal University of Uberlândia, from August to November 2015. Twenty-four pregnant sheep were used, mean body weight 50 kg and approximate 12 months age. The animals were divided into three treatments (Control, palm oil protected fat and cottonseed). The corn silage was used as the bulk. The roughage: concentrate ratio was 60:40 (on dry matter). Measurements of body weight, body condition score and biometric measurements were taken every 21 days. Ingestive behavior evaluation was performed throughout the experiment. The consumption measurement was made in descriptive way. Shortly after birth, the sheep and their offspring were weighed, the barrel circumference was taken, and samples from the mother's blood were collected. The experimental design was completely randomized with repeated measurements in time. It was verified greater time spent with feeding and chewing in the treatment with cottonseed and consequently less time in idle. There was no statistical difference in weight, barrel circumference (BC) and body condition score (BCS) of the sheep under the different treatments. There was a linear growth in BC measurements and sheep weight over the evaluated periods. There were differences between the treatments for the offspring's average weight. The use of cottonseed as a fat source increases the birth weight of the progeny and improves pregnant sheep metabolic profile in relation to inert fat.


Introduction
Ewes in the pregnancy final third show a decrease in feed consumption due to physical limitations of the rumen. It occurs due to the fetus growth as well as it leads to energy demand increase. Thus, if these animals are not fed diets rich in energy, they start to mobilize large amounts of fat and they can develop metabolic problems, such as toxemia of pregnancy (MACEDO JUNIOR, 2008). In this sense, it has been proposed to use diets rich in fats to increase energy consumption in this phase, considered critical within the sheep production system.
Lipids have a higher energy value than any other nutrient, besides being the most important source of energy reserve for animals (NATIONAL RESEARCH CONCIL, 2007). The substitution of non-fibrous carbohydrates for lipids prevents the formation of lactic acid and the high production of volatile fatty acids in the rumen ( VAN SOEST, 1994). However, its addition to the diet can alter rumen fermentation and decrease fiber digestibility (JENKINS et al., 2008), making it necessary to attend to the ether extract (EE) amount in diets for ruminants. In research by Millen et al. (2009), the average EE content was 4.7%. Similar to this result, Oliveira and Millen (2014) observed an average 4.6% EE content in diets, while Pinto and Millen (2016) reported that confinement diets have an average of 5% EE.
Unsaturated fatty acids are considered toxic to the ruminal microbiota. This toxicity may be related to the lipid's ability to disrupt the structure of cell membranes.
However, ruminal microorganisms make use of biohydrogenation, converting unsaturated to saturated fatty acids, placing hydrogen in the carbon chain, and making them less toxic (PALMQUIST & MATTOS, 2006). Therefore, one way of having the benefit of using lipids without damaging the rumen environment and obtaining higher levels of EE in the diet would be to use protected fat.
Another strategy used to reduce the negative effects of lipids on ruminal fermentation is the use of oilseeds, such as cottonseed, once the fatty acid sources of these products are partially protected from ruminal degradation (OLIVEIRA et al., 2011). Bearing that in mind, the way fat is also offered alters the harmful effects of the rumen, since fatty acids in oilseed grains (cottonseed, soybeans, sunflower, peanuts, among others) become less problematic than the direct intake from the oil of these oilseeds, as they are protected by the seed structures and released more slowly (SILVA et al., 2014).
The paper hypothesizes the use of these fat sources in relation to soluble carbohydrate might improve the productive, nutritional, and metabolic performance of pregnant ewes, from the middle and final thirds.
In this context, the objective is to evaluate fat sources with different degradation profiles in pregnant sheep diet on intake, ingestive behavior and productive performance.

Material and methods
The experiment was conducted at Capim Branco experimental farm at  1956). Thus, it was possible to estimate the dry matter intake (DMI) per pen weekly, generating a graph with each pen intake for comparison. Intake was obtained by the difference between the amount of offered feed and subtracted by leftovers after 24 hours.
The biometric measurement of barrel circumference (BC), given in cm, was measured with the aid of a tape measure. The contour of the abdominal cavity is read from the bottom to the ribs where the void of the animal is located. The body condition score (BCS) was performed by a single trained evaluator by palpating the lumbar vertebrae, allowing the assessment of fat deposition in the animal, according to the method described by Russel et al. (1969), in which 1-very thin; 2-lean; 3-normal; 4-fat; 5-obese. WeiHeng® digital scales with an accuracy of 10 grams were used to assess the animals' weight. Biometric measurements and animal weighing were performed at 0, Before birth (approximately five days) the animals were taken to the floor pens. Right after birth, the sheep were weighed with the aid of an electronic scale and after weighing, the following measurements were taken: BCS and BC of the mother and weight of the lamb. Sheep blood was also collected by venipuncture of the jugular with the aid of vacuntainner and tubes without anticoagulant. Soon after, each animal samples were centrifuged, pipetted, and stored in eppendorfs® cryotubes for further laboratory analysis using a commercial kit from Lab Test®.
Data normality and residual tests by W and D statistics (Shapiro & Wilk, 1965;Lilliefors 1967) and homoscedasticity of treatment variances (Levene, 1960) were performed. A completely randomized design was used (this procedure was used to make the experimental groups with initial weight and BCS more homogeneous. However, afterwards, a draw was carried out between the treatments, which guaranteed equal probability of each individual falling into any of treatments) with three treatments: control, palm oil protected fat and cottonseed, and eight repetitions, the periods being evaluated as repeated measures over time.
For the variables in which the sphericity condition was not accepted, we used the mixed model analysis, in which all covariance structures (Σ) available in the SAS software package [(ANTE (1)) -Anti-Dependence; (AR (1)) -First order autoregressive; (ARH (1) Treatments means were evaluated by t-student test, according to criteria established by Sampaio (2002) and the periods by means of linear regression analysis in which the significance of the model (linear or quadratic) was observed, the significance of the coefficients, the nonsignificance of the deviations from linearity and the highest R2, with these first two criteria evaluated at level of significance of 5% probability of type I error. The bays were used as repetition to evaluate dry matter consumption, once there is no repetition of stalls. The body condition score variable, as it is a non-parametric variable, was assessed by using the Kruskal and Wallis (1952)

Control
Palm oil protected fat Cottonseed

Days of pregnancy DMI (Kg/week)
There are some points of oscillation (in all treatments), which is normal, and points of oscillation may be related to the change in the animal's nutritional requirement. According to NRC (2007), sheep pregnancy is divided into two phases: 0-120 days and 120-150 days. At first, fetal growth is exceedingly small, and it does not cause changes in the animal's metabolism, keeping the animal's requirements remarkably close to the maintenance requirements. The last 30 days are marked by fetus exponential growth, which implies major changes in maternal metabolism.
In this phase, the demands on energy, protein, minerals, water, and other nutrients increase significantly, increasing the DMI, which is seen in Graph 1. DMI decreasing near birth is normal and it may be related to fat mobilization from adipose tissue, because of low energy intake once the energy of the diet was not increased close to delivery. It is also associated with uterine volume growth. It is important to highlight that at 123 days of pregnancy, the exact moment of the change suggested by NRC (2007) There was no change in body weight (BW) and body condition score (BCS) as a result of treatments. Body weight increased significantly with advancing pregnancy (Table 2); it is possible due to the increase in DMI at this stage (Graph 1). As previously stated, until 120 days of pregnancy, these animals keep their requirements remarkably close to those of maintenance, which implies low energy expenditure with other activities (fetal growth and milk production), and, therefore, all energy consumed is basically for the animal, reflecting the increase in body weight until 123 days (Table 2). It is expected that pregnancy advancement (after 120 days) these animals will increase body weight, mainly due to the weight increase of the pregnant uterus. Ewes in the final third of pregnancy reach 70% increase in the rate of fetal growth (PILAR et al., 2002). There was a linear increase in barrel circumference (BC) with experimental period advancement ( Table 2) for animals that received the control treatment and palm oil protected fat. Animals that received cottonseed there was a quadratic increase. The increase, in both treatments, is related with period measured (123 days), where fetal growth is low and, therefore, these animals increased their consumption capacity (Graph 1) and consequently increased their body weight at the same period.
There is no statistical difference in relation to average daily gain (ADG) of sheep, indicating that both treatments presented the same ADG throughout the experimental period ( The animals that received cottonseed had a longer time in chewing and eating (Table 3), which can be explained by the seed grain was supplied whole (with linter), allowing for selection or rejection by this ingredient which caused the longest feeding time. Additionally, less idle time was observed in these animals compared with other treatments. Although cottonseed was supplied as a concentrate for these animals, linter (cellulose fibers) contained in it, gives it characteristics of bulky food (> 18% FB, ANDRIGUETTO et al., 1982). The fibers ability to stimulate rumination is called "effectiveness". According to Van Soest (1994), the time spent on rumination is proportional to the cell wall content of bulky foods, with the effectiveness of fiber being a major factor in stimulating chewing, which explains why this treatment has a longer time in chewing than too much. Means followed by different letters in the lines differ (The treatment means were evaluated by the t-student test at the level of significance of 5% probability of type I error); GA: general average; CV: coefficient of variation.
There was no significant effect of fat source on rumination time (Table   3), which may be related to the roughage: concentrate ratio in the diet, which is the same for all treatments (60:40). According to Macedo Junior (2004) the rumination time increases with the increase in total NDF and NDF levels of diets, indicating a close link between the total NDF consumption and the rumination time. The time spent on rumination depends on the diet type and there seems to be little variation in diets rich in grains, reaching a maximum of 10 hours / day in those rich in roughage (QUEIROZ et al., 2001). In this study, the average time spent with rumination was 256.04 minutes / day, equivalent for 4 hours and 16 minutes.
According to Benevides et al. (2011), the time spent on rumination increases with advancing pregnancy. The same authors stated that at 130 days of pregnancy, sheep spent more time ruminating, compared with those at 90 and 110 days, perhaps because in this phase there is better ruminal compression effect due to higher uterus volume uterus and attachments.
Thus, we can conclude that the animals in the present study did not spend much time ruminating, precisely because they are in the transition period (120 days) between two stages of pregnancy. The ability to feed and the feeding behavior are the result of integration amid several factors related to diet, with the environment and the physiological stage of the animal, not being a single factor that controls the intake (BENEVIDES, 2011).
In Table 4 we can see the average weight of the offspring was 4.30 kg, 14% higher than expected according to Ribeiro et al (2008). There were statistical differences amid the treatments for the average weight of the offspring, with lambs born from ewes that received cottonseed compared to those that received palm fat, mainly related to the type of lipid contained in the seed, causing the level to increase cholesterol levels in the same period (Table 6).
Taking the birth weight of 3.77 kg as reference (RIBEIRO et al, 2008), we can conclude that lambs born from sheep treated with cottonseed had higher than expected birth weight (27%), which is related to higher cholesterol level at the farrowing time (Table 6), reflecting greater capacity for energy use by sheep. From the point of view of improving the herd's zootechnical indexes, higher birth weight is advantageous once heavier lambs have a higher survival rate, consequently lower mortality at weaning, higher weight at weaning and in future higher weight at slaughter (GERASSEV et al., 2006;CASTRO et al., 2012). It is important to note that very heavy and large lambs can also present disadvantages, especially at the time of birth, causing dysfunctional births (SCHAFHÄUSER et al., 2004). In that study, the sheep did not present difficulties at the time of farrowing, indicating that the lambs were not too large and heavy to the point of distortion. Means followed by different letters in the lines differ from each other (The treatment means were evaluated by the t-student test at the significance level of 5% probability of type I error); MW: mother's weight; BC: circumference of the mother's barrel; AWO: average weight of the offspring; TWO: total weight of the offspring; GA: general average; CV: coefficient of variation.
For the other variables, there were no statistical differences, being similar for both treatments. Regarding the relationship between the weight of the offspring and their mothers, the greater the relationship, the better, indicating the availability of nutrients was sufficient to cause good offspring growth. Moreover, the average fetal weight corresponded to approximately 7.09% of maternal weight. In some cases, it may generate consumption restrictions due to the space occupied by pregnant uterus in abdominal cavity, which ends up compressing the gastrointestinal tract. Therefore, it may affect the animal's performance, as seen in the present study (Graph 1), once DMI decreased (descriptive data) until birth, demonstrating consumption restriction at this stage.
Regarding protein metabolites at birth, there is statistical difference amid treatments only for the variable creatinine (Table 5). Means followed by different letters in the lines differ from each other (The treatment means were evaluated by the t-student test at the significance level of 5% probability of type I error); * (Varanis, 2018); GA: general average; CV: coefficient of variation; RV: reference values.
The creatinine value was lower in the control and palm oil protected fat treatments, probably because these animals spent more time in leisure (  Means followed by different letters on the different lines between them (As the control media were assessed by the student's test at the 5% significance level of type I error probability); * (Varanis, 2018); ** (Kaneko et al., 2008) indicates that these animals did not develop liver damage. It is noteworthy that the animals supplemented with palm oil protected fat showed the lowest value, probably related to the lipid type (inert). Fat digestion causes the liver to become overloaded, increasing AST level. However, the digestion of inert fat differs somewhat from the "traditional" one, as it passes through the rumen without causing major changes, which resulted in a lower AST value for treatment with protected palm fat.
For the variables alkaline phosphatase and GGT, their general averages are within the reference values (VARANIS, 2018), indicating that animals were efficient from the hepatic point of view and reinforces the result found, indicating that these animals did not develop any liver injury.

Conclusion
Cottonseed usage as a fat source increases the birth weight of the progeny and improves pregnant sheep metabolic profile in relation to inert fat.