Feeding Behavior, Water Intake, and Physiological Parameters of Feedlot Lambs Fed with Diets Containing Babassu Oil Associated with Sunflower Oil Blend

This study aimed to investigate the impact of dietary inclusion of babassu oil (BO) associated with sunflower oil (SO) on feeding behavior, water intake, and physiological parameters of feedlot lambs. Thirty-five castrated male lambs (16.6 kg ± 3.9 kg) were distributed in a randomized block design with 5 treatments (diets) and 7 replications. The tested diets were oil-free diet (OF), 45 g/kg BO (BO), 30 g/kg BO with an additional 15 g/kg SO (1.5 SO), 22.5 g/kg BO with an additional 22.5 g/kg SO (2.25 SO), and 30 g/kg SO with an additional 15 g/kg BO (3.0 SO) on dry matter (DM) basis. The experimental period lasted 60 days. Animals that received BO diet and the combination of BO with SO had lower intakes of DM and neutral detergent fiber (NDF) compared to the control diet (P < 0.05). Differences on the respiratory rate (RR) was observed between animals in the control diet and those in the diets containing SO (P=0.001), with a linear increase in RR as the levels of SO in the diets increased (P=0.004). All physiological parameters showed a time effect (P < 0.05). Animals fed with the control diet had higher water intake via drinking fountain (P=0.030) and total water intake (P=0.029) compared to animals fed with diets containing SO. In relation to SO levels, water intake via drinking fountain (P=0.002), total water intake (P=0.002), and total water intake per kg of DM ingested (P=0.001) linearly increased with the levels increase in the composition of the diets. The tested diets did not alter the feeding behavior of the feedlot lambs. However, the combination of BO with different levels of SO reduced DM and water intake via drinking fountain and RR.


Introduction
Research using alternative sources of feed, such as byproducts derived from biodiesel, has been conducted to reduce production costs and meet the nutritional requirements of animals without harming their productive performance [1].Te use of vegetable oils in the diet of small ruminants can increase the energy density of the diet, reduce fermentation and calorifc increment, and improve the productive efciency of the animals [2].Tis is especially signifcant in tropical regions where high environmental temperatures often negatively impact animal performance in confnement systems [3,4], where animals often need to activate thermoregulatory mechanisms to mitigate stressors, directly afecting feed intake [5].
Vegetable oils are considered highly unsaturated sources and can, therefore, alter the metabolism of the microbial population in the rumen, consequently afecting thermoregulation and the feeding behavior of animals [1,5].Brazil has a great foristic diversity, including a large number of palm species with potential for vegetable oil production.Te babassu palm (Attalea speciosa Mart.ex Spreng) is one of the main native palms in Brazil, primarily found in the states of the Northern and Northeastern regions, predominantly in the Amazon Rainforest region.Babassu is considered the largest global source of wild seed oil, accounting for approximately 72% of the almond weight [6].
Babassu oil (BO) mainly contains medium-chain fatty acids (MCFA), such as lauric acid (47.40%) and myristic acid (15.64%) [7,8].Tis composition of saturated fatty acids is highly relevant in tropical regions, as vegetable oils with a higher proportion of unsaturated fatty acids can be more vulnerable to oxidation [9].Te use of BO in the diet of small ruminants is primarily due to its physicochemical characteristics and its ability to serve as a source of readily available energy, mitigating thermal discomfort in regions with high temperatures [8].Tis was observed by Machado et al. [5], who studied the physiological responses, feeding behavior, and water intake of lambs supplemented with BO or buriti oil in confnement.Tey found that using BO in diets adversely afects the animals' respiratory rate and water intake via drinking fountain, which are important physiological responses to thermal stress in lamb production in tropical regions.
Another vegetable oil widely used in animal feed is sunfower oil (SO).Te sunfower (Helianthus annuus L.) is a plant native to North America and found in Brazil, especially in the south, southeast, midwest [10], and northeast [11] regions.Its seeds yield 52% oil [12], which has a low content of saturated fats (about 10%) [13] and is rich in linoleic acid (25.5-54.9% of total fatty acids).Tis acid can be converted into conjugated linoleic acid (CLA) by ruminants, explaining its high concentration in products derived from these animals [14].
In tropical conditions, the major challenge is to optimize animal performance by reducing the efects of heat stress in small ruminants [15].Dietary efects can infuence the ingestive behavior and physiological responses of animals as they adapt to nutritional changes [16].Terefore, the balance of saturated and unsaturated fatty acids in diets requires comprehensive investigation, as it can mitigate the adverse impacts of vegetable oils when used alone in the ofered feed composition [18,19].In this context, it is important to determine the ideal mixture of vegetable oils to improve animals' feeding behavior and thermoregulatory responses, especially in the face of climatic adversities.
Due to the great importance of nutrition in intensive production systems, research has shown the efciency of babassu oil [7,16,17] and sunfower oil [20][21][22] in the diet of small ruminants, but separately.To the best of our knowledge, no studies have investigated the efect of combining BO with SO on feeding behavior, physiological parameters, and water intake of confned lamb.
Terefore, we hypothesize that combining BO with SO may contribute to the energy density of the diets, promoting a reduction in feeding frequency without afecting the lambs' physiological parameters and water intake.Te aim of this study was to evaluate the dietary association of BO and SO in feedlot lambs, focusing on feeding behavior, water intake, and physiological parameters.

Materials and Methods
2.1.Location.Animal handling followed the guidelines recommended by the Animal Care and Use Committee of the same institution (process number 23115.009213/2019-23).
Te experiment was conducted at the Small Ruminant Sector, Center of Chapadinha Science, Federal University of Maranhão, located in Chapadinha, MA, Brazil (3 °44′26″ S, 43 °21′33″ W, 104 m altitude).Te tropical climate is classifed as "Aw" according to Köppen [23], presenting a hot and rainy season from November to May, with an annual accumulated precipitation of 1670 mm and average annual temperature of 27 °C [24].During the experimental period (60 days), the minimum and maximum temperature and relative air humidity were recorded by a conventional station (Station Code: 82382) of the National Institute of Meteorology [25], obtaining average values of 20.8 °C, 34.9 °C, and 94%, respectively.
Te confnement was carried out in an open shed (without side walls), with a height of 3.5 meters, covered with metal roofng, and with a compacted dirt foor.During the experimental period, data of air temperature (AT) and relative humidity (RH) inside the confnement shed were collected using a digital thermohygrometer (Termômetro Higrometro, INCOTERM-7666.02.0.00,Franca, SP, Brazil).To determine the black globe humidity index (BGHI), the following equation was used [26] (Figure 1): where Tbg � Black globe temperature and Tdp � dew point temperature.
Te Temperature and Humidity Index (THI) (Figure 2) was determined according to Mader et al. [27], where Te thermal comfort/stress ranges observed were classifed according to Silanikove and Koluman [28], who defned THI ranges as follows: <74 (comfortable), 75-79 (moderate stress), 80-85 (stressful), 86-88 (severe stress), and >88 (extreme stress).All animal handling was carried out by the same research team, remaining with the animals since the adaptation period.Terefore, during all analyzes the animals remained calm.  1) were designed to be isonitrogenous and included oil blends with varying proportions of saturated FA (BO) and unsaturated FA (SO), aligned with the treatment specifcations.Te experimental diets were formulated in a roughage-to-concentrate ratio of 30 : 70 to obtain daily gains of 200 g, following the recommendations of the NRC [29].

Animals and Experimental
In the diet formulation process, corn was coarsely ground using a grinder (Trapp, TRF 80, Jaragua do Sul, SC, Brazil) and combined with soybean meal, mineral premix, babassu oil, and/or sunfower oil following the treatments.Te concentrate and Tifton-85 hay were separately weighed utilizing an electronic scale (Welmy, BCW 6/15/30, Santa Barbara d'Oeste, SP, Brazil) and provided as a total mixed ration once daily at 8 a.m.Veterinary Medicine International 2.4.Food Sampling.Te amount of feed ofered and refused was recorded daily to adjust the feed ofered to ensure 10% of refusals.Te animals had access to feed and water ad libitum.Dry matter intake (DMI) and neutral detergent fber intake (NDFI) were obtained by the diference between the total dry matter (DM) of feed ofered and the total DM and total NDF present in leftovers.Samples of the ingredients, diets, and refusals were collected daily and pooled by animal and then frozen at −20 °C for further evaluations.

Feeding Behavior and Water
Intake.Individual observations of the animals were carried out on the 31st and 52nd days of the experiment over 24 hours for better observation of the adaptation of animals to the diets ofered.Te animals' behaviors (feeding, ruminating, idling, and feeder visits; Table 2) were identifed and recorded according to the methodology proposed by Martin and Bateson [30] using instantaneous and continuous sampling, using the focal sampling method and sampling intervals of 10 minute, with continuous periods of 24 h, starting at eight in the morning.Te discretization of the time series was done by counting the discrete periods of feeding, ruminating, and idling.Te average duration of each of the discrete periods was obtained by dividing the daily times of each activity by the number of discrete periods of the same activity [31].
Five trained observers recorded animal behavior data, minimizing interference whenever possible.Each observer was responsible for recording the activities of 7 animals (1 observer per treatment).Te time spent on feeding, ruminating, and idling activities was obtained using digital timers.During the night-time observation data collection, the environment was kept under artifcial lighting.For this, the animals underwent two days of adaptation to the artifcial lighting.
Water was ofered daily, at 7 : 30 am.Water was supplied in 10 L plastic buckets and weighed before delivery and weighed again after 24 h to determine water intake via drinking fountain (WID).Tree buckets containing water were distributed in the shed, close to the stalls, to determine daily evaporation.Water samples were collected for laboratory analysis (Table 3).
For water intake via food (WIF) and total water intake (TWI), the values were calculated according to the following equations: TWI � (water supplied − water evaporated) + diet water.
Total water intake per kg of dry matter ingested (TWI.DMI) was also estimated.
Water intake was calculated as the diference between the daily supply and consumption, adjusted for average daily evaporation.Evaporation was determined by weighing containers flled with a known amount of water from one day to the next.Tese evaluations were conducted over a 5day period and calculated as follows: where WI � water intake; QS � quantity supplied; DO � daily ort; and AE � average daily evaporation.
2.6.Physiological Parameters.Physiological responses of the animals were assessed at 6:00 a.m., 10:00 a.m., 2:00 p.m., and 6:00 p.m. (one minute at each hour) over ten consecutive experimental days (from the 35th to the 44th day).Parameters including the respiratory rate (RR) and body and rectal temperatures (BT and RT, respectively) were measured.Te RR (mov/minute) was determined through direct observation of left fank movements, according to Kawabata et al. [33], counting the number of movements during 15 seconds and the value obtained was multiplied by 4 to determine the RR in movements per minute.Te BT ( °C) was measured using a laser thermometer with an accuracy of ±0.8 °C (Akrom model KR380, Porto Alegre, RS, Brazil), at a distance of 0.50 cm between the animal and observer.Te average values were estimated as the average temperature of the snout, temple, back, and side of the lambs [5].Te RT ( °C) was measured by inserting a veterinary clinical thermometer (Incoterm Termo Med 1.0, São Paulo, SP, Brazil) into the rectum for 2 minutes at a depth of 5 cm so that the bulb was in contact with the animal's mucosa.

Chemical Composition Analysis of Diets.
Te collected samples were ground using a 1 mm Wiley Mill screen (Marconi, Piracicaba, SP, Brazil) to determine the dry matter content (DM; Method 967.01), ash (Method 942.05), ether extract (EE; Method 954.05), and total nitrogen (N; Method 968.06) according to the AOAC [34].Crude protein (CP) was calculated by multiplying the total nitrogen by 6.25.Te neutral detergent fber, assayed with a heat-stable amylase and expressed exclusively of residual ash (aNDFom), was determined following the Mertens [35] method.Te total carbohydrates (TCs) were determined according to Snifen et al. [36] and Mertens [35], respectively.Te nonfber carbohydrates (NFCs) were determined according to Hall [37].Veterinary Medicine International 2.8.Statistical Analysis.Te data were analyzed using the MIXED model procedure in a regression analysis (SAS Institute Inc., Cary, NC), considering the diet as a fxed efect and the animal as a random efect.Bartlett's test was employed to assess the normality and homogeneity of variances for each variable [38].Orthogonal contrasts were used to compare the control group with the diets containing BO or SO.Te means were estimated using the LSMEANS statement, with a signifcance level (α) set at 0.05 for all analyses.Te following mathematical model was used: where Yijk is the response variable for the individual lamb, µ is the overall mean, Di represents the fxed efect of the diet, Ak represents the random efect of the animal, and εijk represents the random error.Physiological parameters were analyzed using the MIXED model procedure (SAS Inst.Inc., Cary, NC) that took into account the same previous efects, with repeated measures over time, as described by the following mathematical model: where µ � the overall mean; Bi � the random efect of block (i � 1-7); Dj � the fxed efect of diet (j � 1-5); Sij � the residual error associated with the animal efect (block × diet); Tk � the fxed efect of time (hours 6, 10, 14, and 18); (DT) jk � the interaction of the diet × time; and Eijk � the residual error.Te most appropriate covariance structure selection was based on Akaike's Information Criteria Corrected (AICC) and Bayesian Information Criteria (BIC).Te best model has the lowest AICC or BIC values.Te covariance matrix that best ft the dataset was a frst-order autoregressive-AR(1) for RR and BT and the frst-order antedependence structure-ANTE(1) for RT.
For the feeding behavior, data were analyzed by PROC GLM procedure (SAS Inst.Inc., Cary, NC) and subjected to analysis of variance at a 5% probability level using the Tukey test.Te following statistical model was used: where µ � the overall mean; Bi � the random efect of block (i � 1-7); Dj � the fxed efect of diet (j � 1-5); and Eij � the residual error.

Feeding Behavior and Water Intake.
Tere was an efect of the diets on DM and NDF intakes (P < 0.05).Animals that received diets containing BO and the combination of BO with SO had lower intakes of DM and NDF, compared to the control diet.Feeding behavior did not show any treatment efects (P > 0.05) when comparing the control diet with the BO diet or the SO diets (Table 4).Animals fed with the control diet had higher WID (P � 0.030) and TWI (P � 0.029) compared to those receiving the SO diets.About SO levels, it was observed that WID (P � 0.002), TWI (P � 0.002), and TWI.DMI (P � 0.001) linearly increased with the levels increase in the composition of the diets.Tere was no efect of the tested diets on WIF (P > 0.05) (Table 4).

Physiological Parameters.
Tere was a signifcant difference in RR between animals on the control diet and those on diets containing SO (P � 0.001), with an increase in RR as the levels of SO in the diets increased (P � 0.004) (Table 5).Lower RR was achieved with the inclusion of 15 g/kg of SO plus 30 g/kg of BO in the tested diets.For variables related to BT and RT, there was no efect of the diets (P > 0.05) (Table 5).All physiological parameters showed an efect of time (P < 0.05) relative to the treatments (Table 5).

Discussion
As expected due to the energy density, diets containing BO and BO and SO combination reduced the animals' DM and NDF intake.However, contrary to the hypothesis raised, diets containing BO and its combination with SO reduced RR, WID, TWI, and TWI.DMI, compared to the control diet.Te composition of the diets may have infuenced the responses obtained by reducing ruminal fermentation and calorifc increment.Terefore, further research is necessary and relevant, involving the development of new diets with diferent levels of association between the tested oils or the use of oils in the composition.

Environmental Variables.
Te animals remained within the thermal comfort zone during the period from 00:00 to 06:00, with THI values below 74.According to Silanikove and Koluman [28], these data indicate thermal comfort.Between 07:00 a.m. and 02:00 p.m. and from 06:00 p.m. onwards, the animals experienced moderate stress [28], with THI values between >74 and 79, and severe stress during the period from 03:00 to 05:00 p.m.When observing BGHI (Figure 2), we can corroborate the fndings for THI, as BGHI values were above 80 between 10:00 a.m. and 05:00 p.m., which is considered above the thermally neutral zone for adult sheep in dryland regions, according to Oliveira et al. [39].Tese factors may have infuenced the animals' physiological responses to environmental variables, indicating a time efect (Tables 5 and 6).Tis underscores the importance of establishing indicators of thermal discomfort to implement measures that mitigate animal stress without compromising productive variables.OF: oil-free diet (control); BO: 45 g/kg BO; 1.5 SO: 30 g/kg BO with an additional 15 g/kg SO; 2.25 SO: 22.5 g/kg BO with an additional 22.5 g/kg SO; 3.0 SO: 30 g/kg SO with an additional 15 g/kg BO; WID: water intake via drinking fountain; WIF: water intake via food; TWI: total water intake; TWI.DMI: Total water intake per kg of dry matter ingested; OF × BO: orthogonal contrast between the oil-free and babassu oil diets; OF × SO: orthogonal contrast between the oil-free and sunfower oil diets; SEM: standard error of mean; L: linear efect; Q: quadratic efect.Signifcant at the 5% probability level.Veterinary Medicine International 4.2.Feeding Behavior and Water Intake.Te lower intake of DMI and NDF by animals fed the BO diet and diets containing SO, compared to the diet without oil, may be related to the fact that these diets were more energy dense.Te control diet contained 2.7% EE, while the other diets contained, on average, 6.9% (Table 1), due to the presence of oils, which may have limited DMI by the animals to meet their nutritional energy needs, generating a satiety efect earlier than the control treatment.Yamamoto et al. [40], studying sources of vegetable oils in the diet of confned lambs, reported that diets with higher energy content limit intake, corroborating with our fndings.
In addition, it should be noted that among the tested diets, the BO diet resulted in the lowest intake of DM and NDF.Tis can be attributed to the nutritional properties of BO, which is rich in medium-chain fatty acids [17,41].Since these fatty acids are saturated and have a lower molecular weight [42], they are absorbed more quickly, increasing the animal's metabolism and leading to lower consumption [43].It is also worth mentioning that lauric acid (C12 : 0) is the predominant saturated fatty acid in babassu oil (18.29%;Table 1), which may have increased the production of cholecystokinin, reducing feed intake by inhibiting gastric emptying, consequently decreasing motility and the rate of digestion through the gastrointestinal compartments [44].On the other hand, the higher acceptability of diets containing SO may be related to its organoleptic characteristics and composition, being rich in polyunsaturated fatty acids.Tese fatty acids undergo the process of ruminal biohydrogenation to become saturated for absorption [45], making them more acceptable to the animals.Te lack of diference in feeding behavior can likely be attributed to the diets' similarity in terms of protein and NDF content and their nutritional adequacy for supporting productive performance.Tese mechanisms can infuence food digestion and its passage rate through the gastrointestinal tract of ruminants.However, animals can adjust their ingestive behavior by modifying one or more components to overcome intake limitations and obtain the required nutrient quantity [46].
Te greater water intake via drinking fountain and the total water intake by animals that received the control diet, compared to animals that received diets with SO, can be attributed to the higher energy concentration in the last ones, combined with the greater intake of dry matter that the animals subjected to this treatment presented.Te lipids allow animals to quickly meet their energy needs due to the high caloric density of the diets [47], which leads to a lower DM intake and, consequently, a lower water intake.Tus, diets containing SO provided a limiting efect on animals, reducing dry matter intake, resulting in lower water intake.4.3.Physiological Parameters.Te higher RR was observed in animals receiving the control diet, likely due to their higher dry matter intake.Diets with high energy content limit dry matter intake, possibly reducing metabolism and lowering respiratory rates [4,48].According to Neiva et al. [49], the diet type signifcantly infuences the susceptibility of the animals to environmental efects, even for hairless breeds originating from tropical regions.Terefore, interactions among the feed type, intake, environment, and physiological parameters should be considered to improve animal performance [48,50].Te animals RR ranged from 34.0 to 44.0 mov/minute, which is above to normal values (20-34 mov/minute) of those established by Reece [51].Tis increase in RR can be justifed to the calorifc increment from fermentation, digestion, absorption, and metabolism, stimulating elevations in this physiological response to maintain homeothermy [52].
Te increase in RR is highly efective for heat dissipation, as the greater volume of air inspired/exhaled by the animal cools and moistens it, leading to increased heat loss through evaporation [53].However, rapid and continuous breathing can add endogenous heat and divert energy that could be used in other metabolic and productive processes [4].Terefore, in environments with high temperatures, the inclusion of 15 g/kg of SO plus 30 g/kg of BO may contribute to animal thermoregulation, as lower RR was observed in animals consuming this diet, indicating reduced endogenous heat production.In addition, this diet resulted in lower water intake for the animals, possibly due to increased metabolic water formation from nutrient oxidation.Tis diet can be recommended for regions with high temperatures and water scarcity.However, the energy expenditure required for the animal to maintain homeothermy may lead to a reduction in daily weight gain [54], requiring additional studies to evaluate the efect of the tested diets on the productive performance of confned lambs, in addition to a greater number of animals so that feeding behavior can be investigated with greater accuracy.
Te minimal increase observed in BT (35.02-35.65 °C) and RT (39.15-39.45 °C) indicates that excess heat was effciently dissipated through the animals' inherent thermoregulatory mechanisms [55].Oliveira et al. [39] studied the efect of diets with high and low energy density on the physiological parameters of lambs and found similar values to this study for rectal temperature (39.16-39.25 °C).However, the authors observed that the lambs had lower body temperatures (31.47-31.98°C) than we observed, which may be related to the location where the experiment was conducted and the environmental conditions (Brazilian Cerrado, which has an Aw type climate-seasonal tropical, characterized by dry winters and rainy summers, with temperatures reaching up to 39 °, and with a thermal sensation of 42 °) [23].

. Conclusions
Te used of BO and SO in lamb's diets reduced DMI and water intake via drinking fountain but did not alter the feeding behavior of the confned animals.Te combination of 30 g/kg BO with 15 g/kg SO was the diet that caused a greater decreased respiratory rate 8 Veterinary Medicine International

Figure 1 :Figure 2 :
Figure 1: Average values of air temperature (AT), relative humidity (RH), and Black Globe Humidity Index (BGHI) obtained per hour during the experimental period (60 days).

Table 1 :
Proportion of the ingredients and chemical composition of the experimental diets.
4F: oil-free diet (control); BO: 45 g/kg BO; 1.5 SO: 30 g/kg BO with an additional 15 g/kg SO; 2.25 SO: 22.5 g/kg BO with an additional 22.5 g/kg SO; 3.0 SO: 30 g/kg SO with an additional 15 g/kg BO; FM: fresh matter; DM: dry matter.4VeterinaryMedicine International WIF � intake of food in natural matter − dry matter intake,

Table 2 :
Description of the variables used to evaluate the feeding behavior of lambs fed with diets containing vegetal oils with diferent fatty acid profle.

Table 3 :
Analysis of the water ofered to the animals during the experimental period.

Table 4 :
Feeding behavior and water intake of lambs fed with diets containing vegetal oils with diferent fatty acid profle.

Table 5 :
Respiratory rate (RR, mov/min), body temperature (BT, °C), and rectal temperature (RT, °C) of lambs fed with diets containing vegetal oils with diferent fatty acid profle.BO: orthogonal contrast between the oil-free and babassu oil diets; OF × SO: orthogonal contrast between the oil-free and sunfower oil diets; SEM: standard error of mean; L: linear efect; Q: quadratic efect; H: time efect.Signifcant at the 5% probability level.

Table 6 :
Efect of diferent observation times on respiratory rate (RR, mov/min), body temperature (BT, °C), and rectal temperature (RT, °C) of lambs fed with diets containing vegetable oils with diferent fatty acid profles.: standard error of mean.Means followed by diferent letters difer from each other using Tukey's test at a 5% probability level. SEM